Add files via upload

TTS Generator - Various interface bugs & filtering options cleaned up.
TTS Generator - TTSDiff now scans generated text and TTS for errors.
TTS Generator - TTSSRT now creates subtitle files for video production e.g. a Youtube video.
Finetune - Now uses a customised tokenizer to deal with Japanese.
Finetune - Pre flight check and warning messages.
Finetune - Extra documentation and warnings.
Entire file structure has been re-organised to simplify management and future changes.
Documentation (built in and Github) has been rewritten/tidied up.
Requirements files have been cleaned up and simplified.
ATsetup has been re-written as necessary with additional options.
DeepSpeed moved up to version 14.
Standalone Application moved to PyTorch 2.2.1.
Nvidia CUDA Toolkit installation is NO LONGER needed (other than to compile DeepSpeed on Linux)

Tested on Linux and Windows.

.github/FUNDING.yml

@@ -0,0 +1 @@
+ko_fi: erew123

.github/ISSUE_TEMPLATE/bug-report-support-request.md

@@ -1,13 +1,14 @@
 ---
-name: Bug report/Support request
+name: Support request OR Bug report
 about: Create a report to help us improve
 title: ''
 labels: ''
 assignees: ''
 
 ---
+🔴 If you have installed AllTalk in a custom Python environment, I will only be able to provide limited assistance/support. AllTalk draws on a variety of scripts and libraries that are not written or managed by myself, and they may fail, error or give strange results in custom built python environments.
 
-Please generate a diagnostics report and upload the "diagnostics.log".
+🔴 Please generate a diagnostics report and upload the "diagnostics.log" as this helps me understand your configuration.
 
 https://github.com/erew123/alltalk_tts/tree/main?#-how-to-make-a-diagnostics-report-file
 

.github/workflows/publish-docker.yml

@@ -0,0 +1,67 @@
+name: Publish Docker image
+
+on:
+  push:
+    branches: [main]
+  release:
+    types: [published]
+  workflow_dispatch:
+
+jobs:
+  build_and_push:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        include:
+          - dockerfile: "Dockerfile"
+            platforms: "linux/amd64"
+            tag_name: "cpu"
+          - dockerfile: "nvidia.Dockerfile"
+            platforms: "linux/amd64"
+            tag_name: "cuda"
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Lowercase Actor and Extract Repo Name
+        run: |
+          echo "ACTOR_LOWER=$(echo ${{ github.actor }} | tr '[:upper:]' '[:lower:]')" >> $GITHUB_ENV
+          echo "REPO_NAME=$(echo ${{ github.repository }} | cut -d'/' -f 2)" >> $GITHUB_ENV
+
+      - name: Login to GitHub Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ env.ACTOR_LOWER }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Login to DockerHub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ vars.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          file: ./${{ matrix.dockerfile }}
+          platforms: ${{ matrix.platforms }}
+          push: true
+          tags: |
+            ghcr.io/${{ env.ACTOR_LOWER }}/${{ env.REPO_NAME }}:${{ matrix.tag_name }}-${{ github.sha }}
+            ghcr.io/${{ env.ACTOR_LOWER }}/${{ env.REPO_NAME }}:${{ matrix.tag_name }}
+            ghcr.io/${{ env.ACTOR_LOWER }}/${{ env.REPO_NAME }}:latest
+            ${{ vars.DOCKERHUB_USERNAME }}/${{ env.REPO_NAME }}:${{ matrix.tag_name }}-${{ github.sha }}
+            ${{ vars.DOCKERHUB_USERNAME }}/${{ env.REPO_NAME }}:${{ matrix.tag_name }}
+            ${{ vars.DOCKERHUB_USERNAME }}/${{ env.REPO_NAME }}:latest
+
+      - name: Docker Hub Description
+        uses: peter-evans/dockerhub-description@v3.4.2
+        with:
+          username: ${{ vars.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+          repository: ${{ vars.DOCKERHUB_USERNAME }}/${{ env.REPO_NAME }}

Deepspeed-Windows/11.1/csrc/quantization/pt_binding.cpp

@@ -1,285 +0,0 @@
-// Copyright (c) Microsoft Corporation.
-// SPDX-License-Identifier: Apache-2.0
-
-// DeepSpeed Team
-
-#include <ATen/cuda/CUDAContext.h>
-#include <torch/extension.h>
-#include <cassert>
-#include <vector>
-#include "quantization.h"
-
-template <typename T>
-at::Tensor ds_quantize(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if ((((size / groups) - 1) / 4096 + 1) <= 256) {
-        launch_fake_quantize_kernel(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-template <typename T>
-at::Tensor ds_sr_quantize(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if (((size / groups) / 4 / 1024) <= 256) {
-        launch_sr_fake_quantize_kernel(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-template <typename T>
-at::Tensor ds_quantize_asym(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if ((((size / groups) - 1) / 4096 + 1) <= 256) {
-        launch_fake_quantize_kernel_asym(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-template <typename T>
-at::Tensor ds_sr_quantize_asym(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if (((size / groups) / 4 / 1024) <= 256) {
-        launch_sr_fake_quantize_kernel_asym(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-std::vector<at::Tensor> quantize_kernel(at::Tensor& input_vals,
-                                        int groups,
-                                        int numBits,
-                                        quantize::Type quantType)
-{
-    auto dtype = at::kFloat;
-    auto params_options = at::TensorOptions()
-                              .dtype(dtype)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-    const int param_elems = (quantize::requires_offset(quantType)) ? 2 : 1;
-    auto params = torch::empty({groups, param_elems}, params_options);
-
-    auto output_options = at::TensorOptions()
-                              .dtype(at::kChar)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    auto output_sizes = input_vals.sizes().vec();
-    output_sizes[output_sizes.size() - 1] /= numBits == 8 ? 1 : 2;
-    auto output = torch::empty(output_sizes, output_options);
-
-    const int elems_per_group = at::numel(input_vals) / groups;
-
-    launch_quant((int8_t*)output.data_ptr(),
-                 (float*)params.data_ptr(),
-                 (__half*)input_vals.data_ptr(),
-                 groups,
-                 elems_per_group,
-                 numBits,
-                 quantType,
-                 at::cuda::getCurrentCUDAStream());
-
-    return {output, params};
-}
-
-template <typename T>
-at::Tensor dequantize(at::Tensor& quantized_data,
-                      at::Tensor& params,
-                      int groups,
-                      int num_bits,
-                      quantize::Type quant_type)
-{
-    auto dtype = (std::is_same<T, float>::value) ? torch::kFloat32 : torch::kFloat16;
-    auto output_options = at::TensorOptions()
-                              .dtype(dtype)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    auto output_sizes = quantized_data.sizes().vec();
-    output_sizes[output_sizes.size() - 1] *= num_bits == 8 ? 1 : 2;
-    auto output = torch::empty(output_sizes, output_options);
-
-    const int total_elems = at::numel(output);
-    const int elems_per_group = total_elems / groups;
-
-    launch_dequantize_kernel((T*)output.data_ptr(),
-                             (const int8_t*)quantized_data.data_ptr(),
-                             (const float*)params.data_ptr(),
-                             quant_type,
-                             num_bits,
-                             elems_per_group,
-                             total_elems,
-                             at::cuda::getCurrentCUDAStream());
-
-    return output;
-}
-
-at::Tensor dequantize_int4_to_half_experimental(at::Tensor& data_in,
-                                                at::Tensor& scale_buffer,
-                                                at::Tensor& min_val_buffer,
-                                                int num_group,
-                                                int group_size)
-{
-    auto output_options = at::TensorOptions().dtype(at::kHalf).device(at::kCUDA);
-    auto output = torch::empty({num_group, group_size}, output_options);
-
-    launch_dequantize_int4_to_half_experimental((uint8_t*)data_in.data_ptr(),
-                                                (half*)output.data_ptr(),
-                                                (half*)scale_buffer.data_ptr(),
-                                                (half*)min_val_buffer.data_ptr(),
-                                                num_group,
-                                                group_size,
-                                                at::cuda::getCurrentCUDAStream());
-
-    return output;
-}
-
-std::vector<at::Tensor> ds_swizzle_quant(at::Tensor& input_vals,
-                                         int groups,
-                                         int num_bits,
-                                         quantize::Type quant_type,
-                                         int pipeline_size,
-                                         int nodes,
-                                         int devices_per_node)
-{
-    auto scales_options = at::TensorOptions()
-                              .dtype(at::kFloat)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-    const int scales_elems = (quantize::requires_offset(quant_type)) ? 2 : 1;
-    auto scales = torch::empty({groups, scales_elems}, scales_options);
-
-    auto output_options = at::TensorOptions()
-                              .dtype(at::kChar)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    const int quantization_scalar = 8 / num_bits;
-    const int compressed_vals = at::numel(input_vals) / quantization_scalar;
-
-    auto output = torch::empty({compressed_vals}, output_options);
-    const int elems_per_group = at::numel(input_vals) / groups;
-
-    launch_swizzled_quant((int8_t*)output.data_ptr(),
-                          (float*)scales.data_ptr(),
-                          (__half*)input_vals.data_ptr(),
-                          num_bits,
-                          quant_type,
-                          groups,
-                          elems_per_group,
-                          pipeline_size,
-                          nodes,
-                          devices_per_node,
-                          at::cuda::getCurrentCUDAStream());
-
-    return {output, scales};
-}
-
-std::vector<at::Tensor> quantized_reduction(at::Tensor& input_vals,
-                                            at::Tensor& input_scales,
-                                            int in_groups,
-                                            int out_groups,
-                                            int num_bits,
-                                            quantize::Type quant_type,
-                                            int devices_per_node)
-{
-    auto scales_options = at::TensorOptions()
-                              .dtype(at::kFloat)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-    const int scales_elems = (quantize::requires_offset(quant_type)) ? 2 : 1;
-    auto scales = torch::empty({out_groups, scales_elems}, scales_options);
-
-    auto output_options = at::TensorOptions()
-                              .dtype(at::kChar)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-     std::vector<int64_t> sz_vector(input_vals.sizes().begin(), input_vals.sizes().end());
-     sz_vector[sz_vector.size() - 1] = sz_vector.back() / devices_per_node;  // num of GPU per nodes
-     at::IntArrayRef sz(sz_vector);
-     auto output = torch::empty(sz, output_options);
-
-     const int elems_per_in_tensor = at::numel(input_vals) / devices_per_node;
-     const int elems_per_in_group = elems_per_in_tensor / (in_groups / devices_per_node);
-     const int elems_per_out_group = elems_per_in_tensor / out_groups;
-
-// old code:
-//    std::vector<long int> sz(input_vals.sizes().begin(), input_vals.sizes().end());
-//    sz[sz.size() - 1] = sz.back() / devices_per_node;  // num of GPU per nodes
-//    const int elems_per_in_tensor = at::numel(input_vals) / devices_per_node;
-//    auto output = torch::empty(sz, output_options);
-
-//    const int elems_per_in_group = elems_per_in_tensor / (in_groups / devices_per_node);
-//    const int elems_per_out_group = elems_per_in_tensor / out_groups;
-
-    launch_dequant_reduce((int8_t*)output.data_ptr(),
-                          (float*)scales.data_ptr(),
-                          (const int8_t*)input_vals.data_ptr(),
-                          (const float*)input_scales.data_ptr(),
-                          devices_per_node,
-                          num_bits,
-                          quant_type,
-                          out_groups,
-                          elems_per_out_group,
-                          elems_per_in_tensor,
-                          in_groups / devices_per_node,
-                          elems_per_in_group,
-                          at::cuda::getCurrentCUDAStream());
-    return {output, scales};
-}
-
-PYBIND11_MODULE(TORCH_EXTENSION_NAME, m)
-{
-    m.def("ds_quantize_fp32", &ds_quantize<float>, "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def("ds_quantize_fp16", &ds_quantize<__half>, "DeepSpeed Quantize with fp16 (CUDA)");
-    m.def("ds_sr_quantize_fp32", &ds_sr_quantize<float>, "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def("ds_sr_quantize_fp16", &ds_sr_quantize<__half>, "DeepSpeed Quantize with fp16 (CUDA)");
-    m.def("ds_quantize_asym_fp32", &ds_quantize_asym<float>, "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def(
-        "ds_quantize_asym_fp16", &ds_quantize_asym<__half>, "DeepSpeed Quantize with fp16 (CUDA)");
-    m.def("ds_sr_quantize_asym_fp32",
-          &ds_sr_quantize_asym<float>,
-          "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def("ds_sr_quantize_asym_fp16",
-          &ds_sr_quantize_asym<__half>,
-          "DeepSpeed Quantize with fp16 (CUDA)");
-    pybind11::enum_<quantize::Type>(m, "QuantizationType")
-        .value("Symmetric", quantize::Type::Symmetric)
-        .value("Asymmetric", quantize::Type::Asymmetric)
-        .export_values();
-    m.def("quantize", &quantize_kernel);
-    m.def("dequantize", &dequantize<__half>);
-    m.def("dequantize_fp32", &dequantize<float>);
-    m.def("dequantize_int4_to_half_experimental",
-          &dequantize_int4_to_half_experimental,
-          "Dequantize int4 to half (experimental)");
-    m.def("swizzle_quant", &ds_swizzle_quant);
-    m.def("quantized_reduction", &quantized_reduction);
-}

Deepspeed-Windows/11.1/deepspeed/env_report.py

@@ -1,160 +0,0 @@
-# Copyright (c) Microsoft Corporation.
-# SPDX-License-Identifier: Apache-2.0
-
-# DeepSpeed Team
-
-import os
-import torch
-import deepspeed
-import subprocess
-import argparse
-import psutil
-from .ops.op_builder.all_ops import ALL_OPS
-from .git_version_info import installed_ops, torch_info
-from deepspeed.accelerator import get_accelerator
-
-GREEN = '\033[92m'
-RED = '\033[91m'
-YELLOW = '\033[93m'
-END = '\033[0m'
-SUCCESS = f"{GREEN} [SUCCESS] {END}"
-OKAY = f"{GREEN}[OKAY]{END}"
-WARNING = f"{YELLOW}[WARNING]{END}"
-FAIL = f'{RED}[FAIL]{END}'
-INFO = '[INFO]'
-
-color_len = len(GREEN) + len(END)
-okay = f"{GREEN}[OKAY]{END}"
-warning = f"{YELLOW}[WARNING]{END}"
-
-
-def op_report(verbose=True):
-    max_dots = 23
-    max_dots2 = 11
-    h = ["op name", "installed", "compatible"]
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    print("DeepSpeed C++/CUDA extension op report")
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-
-    print("NOTE: Ops not installed will be just-in-time (JIT) compiled at\n"
-          "      runtime if needed. Op compatibility means that your system\n"
-          "      meet the required dependencies to JIT install the op.")
-
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    print("JIT compiled ops requires ninja")
-    ninja_status = OKAY if ninja_installed() else FAIL
-    print('ninja', "." * (max_dots - 5), ninja_status)
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    print(h[0], "." * (max_dots - len(h[0])), h[1], "." * (max_dots2 - len(h[1])), h[2])
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    installed = f"{GREEN}[YES]{END}"
-    no = f"{YELLOW}[NO]{END}"
-    for op_name, builder in ALL_OPS.items():
-        dots = "." * (max_dots - len(op_name))
-        is_compatible = OKAY if builder.is_compatible(verbose) else no
-        is_installed = installed if installed_ops.get(op_name, False) else no
-        dots2 = '.' * ((len(h[1]) + (max_dots2 - len(h[1]))) - (len(is_installed) - color_len))
-        print(op_name, dots, is_installed, dots2, is_compatible)
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-
-
-def ninja_installed():
-    try:
-        import ninja  # noqa: F401 # type: ignore
-    except ImportError:
-        return False
-    return True
-
-
-def nvcc_version():
-    import torch.utils.cpp_extension
-    cuda_home = torch.utils.cpp_extension.CUDA_HOME
-    if cuda_home is None:
-        return f"{RED} [FAIL] cannot find CUDA_HOME via torch.utils.cpp_extension.CUDA_HOME={torch.utils.cpp_extension.CUDA_HOME} {END}"
-    try:
-        output = subprocess.check_output([cuda_home + "/bin/nvcc", "-V"], universal_newlines=True)
-    except FileNotFoundError:
-        return f"{RED} [FAIL] nvcc missing {END}"
-    output_split = output.split()
-    release_idx = output_split.index("release")
-    release = output_split[release_idx + 1].replace(',', '').split(".")
-    return ".".join(release)
-
-
-def get_shm_size():
-    try:
-        temp_dir = os.getenv('TEMP') or os.getenv('TMP') or os.path.join(os.path.expanduser('~'), 'tmp')
-        shm_stats = psutil.disk_usage(temp_dir)
-        shm_size = shm_stats.total
-        shm_hbytes = human_readable_size(shm_size)
-        warn = []
-        if shm_size < 512 * 1024**2:
-            warn.append(
-                f" {YELLOW} [WARNING] Shared memory size might be too small, consider increasing it. {END}"
-            )
-            # Add additional warnings specific to your use case if needed.
-        return shm_hbytes, warn
-    except Exception as e:
-        return "UNKNOWN", [f"Error getting shared memory size: {e}"]
-
-def human_readable_size(size):
-    units = ['B', 'KB', 'MB', 'GB', 'TB']
-    i = 0
-    while size >= 1024 and i < len(units) - 1:
-        size /= 1024
-        i += 1
-    return f'{size:.2f} {units[i]}'
-
-
-def debug_report():
-    max_dots = 33
-
-    report = [("torch install path", torch.__path__), ("torch version", torch.__version__),
-              ("deepspeed install path", deepspeed.__path__),
-              ("deepspeed info", f"{deepspeed.__version__}, {deepspeed.__git_hash__}, {deepspeed.__git_branch__}")]
-    if get_accelerator().device_name() == 'cuda':
-        hip_version = getattr(torch.version, "hip", None)
-        report.extend([("torch cuda version", torch.version.cuda), ("torch hip version", hip_version),
-                       ("nvcc version", (None if hip_version else nvcc_version())),
-                       ("deepspeed wheel compiled w.", f"torch {torch_info['version']}, " +
-                        (f"hip {torch_info['hip_version']}" if hip_version else f"cuda {torch_info['cuda_version']}"))
-                       ])
-    else:
-        report.extend([("deepspeed wheel compiled w.", f"torch {torch_info['version']} ")])
-
-    report.append(("shared memory (/dev/shm) size", get_shm_size()))
-
-    print("DeepSpeed general environment info:")
-    for name, value in report:
-        warns = []
-        if isinstance(value, tuple):
-            value, warns = value
-        print(name, "." * (max_dots - len(name)), value)
-        if warns:
-            for warn in warns:
-                print(warn)
-
-
-def parse_arguments():
-    parser = argparse.ArgumentParser()
-    parser.add_argument('--hide_operator_status',
-                        action='store_true',
-                        help='Suppress display of installation and compatibility statuses of DeepSpeed operators. ')
-    parser.add_argument('--hide_errors_and_warnings', action='store_true', help='Suppress warning and error messages.')
-    args = parser.parse_args()
-    return args
-
-
-def main(hide_operator_status=False, hide_errors_and_warnings=False):
-    if not hide_operator_status:
-        op_report(verbose=not hide_errors_and_warnings)
-    debug_report()
-
-
-def cli_main():
-    args = parse_arguments()
-    main(hide_operator_status=args.hide_operator_status, hide_errors_and_warnings=args.hide_errors_and_warnings)
-
-
-if __name__ == "__main__":
-    main()

Deepspeed-Windows/11.2/csrc/quantization/pt_binding.cpp

@@ -1,299 +0,0 @@
-// Copyright (c) Microsoft Corporation.
-// SPDX-License-Identifier: Apache-2.0
-
-// DeepSpeed Team
-
-#include <ATen/cuda/CUDAContext.h>
-#include <torch/extension.h>
-#include <cassert>
-#include <vector>
-#include "quantization.h"
-
-template <typename T>
-at::Tensor ds_quantize(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if ((((size / groups) - 1) / 4096 + 1) <= 256) {
-        launch_fake_quantize_kernel(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-template <typename T>
-at::Tensor ds_sr_quantize(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if (((size / groups) / 4 / 1024) <= 256) {
-        launch_sr_fake_quantize_kernel(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-template <typename T>
-at::Tensor ds_quantize_asym(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if ((((size / groups) - 1) / 4096 + 1) <= 256) {
-        launch_fake_quantize_kernel_asym(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-template <typename T>
-at::Tensor ds_sr_quantize_asym(at::Tensor& vals, int groups, int bits)
-{
-    auto t_size = vals.sizes();
-    int size = 1;
-    for (auto dim : t_size) size *= dim;
-
-    if (((size / groups) / 4 / 1024) <= 256) {
-        launch_sr_fake_quantize_kernel_asym(
-            (T*)vals.data_ptr(), size, groups, bits, at::cuda::getCurrentCUDAStream());
-    }
-    return vals;
-}
-
-std::vector<at::Tensor> quantize_kernel(at::Tensor& input_vals,
-                                        int groups,
-                                        int numBits,
-                                        quantize::Type quantType)
-{
-    auto dtype = at::kFloat;
-    auto params_options = at::TensorOptions()
-                              .dtype(dtype)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-    const int param_elems = (quantize::requires_offset(quantType)) ? 2 : 1;
-    auto params = torch::empty({groups, param_elems}, params_options);
-
-    auto output_options = at::TensorOptions()
-                              .dtype(at::kChar)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    auto output_sizes = input_vals.sizes().vec();
-    output_sizes[output_sizes.size() - 1] /= numBits == 8 ? 1 : 2;
-    auto output = torch::empty(output_sizes, output_options);
-
-    const int elems_per_group = at::numel(input_vals) / groups;
-
-    launch_quant((int8_t*)output.data_ptr(),
-                 (float*)params.data_ptr(),
-                 (__half*)input_vals.data_ptr(),
-                 groups,
-                 elems_per_group,
-                 numBits,
-                 quantType,
-                 at::cuda::getCurrentCUDAStream());
-
-    return {output, params};
-}
-
-template <typename T>
-at::Tensor dequantize(at::Tensor& quantized_data,
-                      at::Tensor& params,
-                      int groups,
-                      int num_bits,
-                      quantize::Type quant_type)
-{
-    auto dtype = (std::is_same<T, float>::value) ? torch::kFloat32 : torch::kFloat16;
-    auto output_options = at::TensorOptions()
-                              .dtype(dtype)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    auto output_sizes = quantized_data.sizes().vec();
-    output_sizes[output_sizes.size() - 1] *= num_bits == 8 ? 1 : 2;
-    auto output = torch::empty(output_sizes, output_options);
-
-    const int total_elems = at::numel(output);
-    const int elems_per_group = total_elems / groups;
-
-    launch_dequantize_kernel((T*)output.data_ptr(),
-                             (const int8_t*)quantized_data.data_ptr(),
-                             (const float*)params.data_ptr(),
-                             quant_type,
-                             num_bits,
-                             elems_per_group,
-                             total_elems,
-                             at::cuda::getCurrentCUDAStream());
-
-    return output;
-}
-
-at::Tensor dequantize_int4_to_half_experimental(at::Tensor& data_in,
-                                                at::Tensor& scale_buffer,
-                                                at::Tensor& min_val_buffer,
-                                                int num_group,
-                                                int group_size)
-{
-    auto output_options = at::TensorOptions().dtype(at::kHalf).device(at::kCUDA);
-    auto output = torch::empty({num_group, group_size}, output_options);
-
-    launch_dequantize_int4_to_half_experimental((uint8_t*)data_in.data_ptr(),
-                                                (half*)output.data_ptr(),
-                                                (half*)scale_buffer.data_ptr(),
-                                                (half*)min_val_buffer.data_ptr(),
-                                                num_group,
-                                                group_size,
-                                                at::cuda::getCurrentCUDAStream());
-
-    return output;
-}
-
-at::Tensor dequantize_int8_to_half_experimental(at::Tensor& data_in,
-                                                at::Tensor& scale_buffer,
-                                                at::Tensor& min_val_buffer,
-                                                int num_group,
-                                                int group_size)
-{
-    auto output_options = at::TensorOptions().dtype(at::kHalf).device(at::kCUDA);
-    auto output = torch::empty({num_group, group_size}, output_options);
-
-    launch_dequantize_int8_to_half_experimental((uint8_t*)data_in.data_ptr(),
-                                                (half*)output.data_ptr(),
-                                                (half*)scale_buffer.data_ptr(),
-                                                (half*)min_val_buffer.data_ptr(),
-                                                num_group,
-                                                group_size,
-                                                at::cuda::getCurrentCUDAStream());
-
-    return output;
-}
-
-std::vector<at::Tensor> ds_swizzle_quant(at::Tensor& input_vals,
-                                         int groups,
-                                         int num_bits,
-                                         quantize::Type quant_type,
-                                         int pipeline_size,
-                                         int nodes,
-                                         int devices_per_node)
-{
-    auto scales_options = at::TensorOptions()
-                              .dtype(at::kFloat)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-    const int scales_elems = (quantize::requires_offset(quant_type)) ? 2 : 1;
-    auto scales = torch::empty({groups, scales_elems}, scales_options);
-
-    auto output_options = at::TensorOptions()
-                              .dtype(at::kChar)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    const int quantization_scalar = 8 / num_bits;
-    const int compressed_vals = at::numel(input_vals) / quantization_scalar;
-
-    auto output = torch::empty({compressed_vals}, output_options);
-    const int elems_per_group = at::numel(input_vals) / groups;
-
-    launch_swizzled_quant((int8_t*)output.data_ptr(),
-                          (float*)scales.data_ptr(),
-                          (__half*)input_vals.data_ptr(),
-                          num_bits,
-                          quant_type,
-                          groups,
-                          elems_per_group,
-                          pipeline_size,
-                          nodes,
-                          devices_per_node,
-                          at::cuda::getCurrentCUDAStream());
-
-    return {output, scales};
-}
-
-std::vector<at::Tensor> quantized_reduction(at::Tensor& input_vals,
-                                            at::Tensor& input_scales,
-                                            int in_groups,
-                                            int out_groups,
-                                            int num_bits,
-                                            quantize::Type quant_type,
-                                            int devices_per_node)
-{
-    auto scales_options = at::TensorOptions()
-                              .dtype(at::kFloat)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-    const int scales_elems = (quantize::requires_offset(quant_type)) ? 2 : 1;
-    auto scales = torch::empty({out_groups, scales_elems}, scales_options);
-
-    auto output_options = at::TensorOptions()
-                              .dtype(at::kChar)
-                              .layout(at::kStrided)
-                              .device(at::kCUDA)
-                              .requires_grad(false);
-
-    std::vector<int64_t> sz_vector(input_vals.sizes().begin(), input_vals.sizes().end());
-    sz_vector[sz_vector.size() - 1] = sz_vector.back() / devices_per_node;  // num of GPU per nodes
-    at::IntArrayRef sz(sz_vector);
-    auto output = torch::empty(sz, output_options);
-
-    const int elems_per_in_tensor = at::numel(input_vals) / devices_per_node;
-    const int elems_per_in_group = elems_per_in_tensor / (in_groups / devices_per_node);
-    const int elems_per_out_group = elems_per_in_tensor / out_groups;
-
-    launch_dequant_reduce((int8_t*)output.data_ptr(),
-                          (float*)scales.data_ptr(),
-                          (const int8_t*)input_vals.data_ptr(),
-                          (const float*)input_scales.data_ptr(),
-                          devices_per_node,
-                          num_bits,
-                          quant_type,
-                          out_groups,
-                          elems_per_out_group,
-                          elems_per_in_tensor,
-                          in_groups / devices_per_node,
-                          elems_per_in_group,
-                          at::cuda::getCurrentCUDAStream());
-    return {output, scales};
-}
-
-PYBIND11_MODULE(TORCH_EXTENSION_NAME, m)
-{
-    m.def("ds_quantize_fp32", &ds_quantize<float>, "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def("ds_quantize_fp16", &ds_quantize<__half>, "DeepSpeed Quantize with fp16 (CUDA)");
-    m.def("ds_sr_quantize_fp32", &ds_sr_quantize<float>, "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def("ds_sr_quantize_fp16", &ds_sr_quantize<__half>, "DeepSpeed Quantize with fp16 (CUDA)");
-    m.def("ds_quantize_asym_fp32", &ds_quantize_asym<float>, "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def(
-        "ds_quantize_asym_fp16", &ds_quantize_asym<__half>, "DeepSpeed Quantize with fp16 (CUDA)");
-    m.def("ds_sr_quantize_asym_fp32",
-          &ds_sr_quantize_asym<float>,
-          "DeepSpeed Quantize with fp32 (CUDA)");
-    m.def("ds_sr_quantize_asym_fp16",
-          &ds_sr_quantize_asym<__half>,
-          "DeepSpeed Quantize with fp16 (CUDA)");
-    pybind11::enum_<quantize::Type>(m, "QuantizationType")
-        .value("Symmetric", quantize::Type::Symmetric)
-        .value("Asymmetric", quantize::Type::Asymmetric)
-        .export_values();
-    m.def("quantize", &quantize_kernel);
-    m.def("dequantize", &dequantize<__half>);
-    m.def("dequantize_fp32", &dequantize<float>);
-    m.def("dequantize_int4_to_half_experimental",
-          &dequantize_int4_to_half_experimental,
-          "Dequantize int4 to half (experimental)");
-    m.def("dequantize_int8_to_half_experimental",
-          &dequantize_int8_to_half_experimental,
-          "Dequantize int8 to half (experimental)");
-    m.def("swizzle_quant", &ds_swizzle_quant);
-    m.def("quantized_reduction", &quantized_reduction);
-}

Deepspeed-Windows/11.2/deepspeed/env_report.py

@@ -1,161 +0,0 @@
-# Copyright (c) Microsoft Corporation.
-# SPDX-License-Identifier: Apache-2.0
-
-# DeepSpeed Team
-
-import os
-import torch
-import deepspeed
-import subprocess
-import argparse
-import psutil
-from .ops.op_builder.all_ops import ALL_OPS
-from .git_version_info import installed_ops, torch_info
-from deepspeed.accelerator import get_accelerator
-
-GREEN = '\033[92m'
-RED = '\033[91m'
-YELLOW = '\033[93m'
-END = '\033[0m'
-SUCCESS = f"{GREEN} [SUCCESS] {END}"
-OKAY = f"{GREEN}[OKAY]{END}"
-WARNING = f"{YELLOW}[WARNING]{END}"
-FAIL = f'{RED}[FAIL]{END}'
-INFO = '[INFO]'
-
-color_len = len(GREEN) + len(END)
-okay = f"{GREEN}[OKAY]{END}"
-warning = f"{YELLOW}[WARNING]{END}"
-
-
-def op_report(verbose=True):
-    max_dots = 23
-    max_dots2 = 11
-    h = ["op name", "installed", "compatible"]
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    print("DeepSpeed C++/CUDA extension op report")
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-
-    print("NOTE: Ops not installed will be just-in-time (JIT) compiled at\n"
-          "      runtime if needed. Op compatibility means that your system\n"
-          "      meet the required dependencies to JIT install the op.")
-
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    print("JIT compiled ops requires ninja")
-    ninja_status = OKAY if ninja_installed() else FAIL
-    print('ninja', "." * (max_dots - 5), ninja_status)
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    print(h[0], "." * (max_dots - len(h[0])), h[1], "." * (max_dots2 - len(h[1])), h[2])
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-    installed = f"{GREEN}[YES]{END}"
-    no = f"{YELLOW}[NO]{END}"
-    for op_name, builder in ALL_OPS.items():
-        dots = "." * (max_dots - len(op_name))
-        is_compatible = OKAY if builder.is_compatible(verbose) else no
-        is_installed = installed if installed_ops.get(op_name, False) else no
-        dots2 = '.' * ((len(h[1]) + (max_dots2 - len(h[1]))) - (len(is_installed) - color_len))
-        print(op_name, dots, is_installed, dots2, is_compatible)
-    print("-" * (max_dots + max_dots2 + len(h[0]) + len(h[1])))
-
-
-def ninja_installed():
-    try:
-        import ninja  # noqa: F401 # type: ignore
-    except ImportError:
-        return False
-    return True
-
-
-def nvcc_version():
-    import torch.utils.cpp_extension
-    cuda_home = torch.utils.cpp_extension.CUDA_HOME
-    if cuda_home is None:
-        return f"{RED} [FAIL] cannot find CUDA_HOME via torch.utils.cpp_extension.CUDA_HOME={torch.utils.cpp_extension.CUDA_HOME} {END}"
-    try:
-        output = subprocess.check_output([cuda_home + "/bin/nvcc", "-V"], universal_newlines=True)
-    except FileNotFoundError:
-        return f"{RED} [FAIL] nvcc missing {END}"
-    output_split = output.split()
-    release_idx = output_split.index("release")
-    release = output_split[release_idx + 1].replace(',', '').split(".")
-    return ".".join(release)
-
-
-def get_shm_size():
-    try:
-        temp_dir = os.getenv('TEMP') or os.getenv('TMP') or os.path.join(os.path.expanduser('~'), 'tmp')
-        shm_stats = psutil.disk_usage(temp_dir)
-        shm_size = shm_stats.total
-        shm_hbytes = human_readable_size(shm_size)
-        warn = []
-        if shm_size < 512 * 1024**2:
-            warn.append(
-                f" {YELLOW} [WARNING] Shared memory size might be too small, consider increasing it. {END}"
-            )
-            # Add additional warnings specific to your use case if needed.
-        return shm_hbytes, warn
-    except Exception as e:
-        return "UNKNOWN", [f"Error getting shared memory size: {e}"]
-
-
-def human_readable_size(size):
-    units = ['B', 'KB', 'MB', 'GB', 'TB']
-    i = 0
-    while size >= 1024 and i < len(units) - 1:
-        size /= 1024
-        i += 1
-    return f'{size:.2f} {units[i]}'
-
-
-def debug_report():
-    max_dots = 33
-
-    report = [("torch install path", torch.__path__), ("torch version", torch.__version__),
-              ("deepspeed install path", deepspeed.__path__),
-              ("deepspeed info", f"{deepspeed.__version__}, {deepspeed.__git_hash__}, {deepspeed.__git_branch__}")]
-    if get_accelerator().device_name() == 'cuda':
-        hip_version = getattr(torch.version, "hip", None)
-        report.extend([("torch cuda version", torch.version.cuda), ("torch hip version", hip_version),
-                       ("nvcc version", (None if hip_version else nvcc_version())),
-                       ("deepspeed wheel compiled w.", f"torch {torch_info['version']}, " +
-                        (f"hip {torch_info['hip_version']}" if hip_version else f"cuda {torch_info['cuda_version']}"))
-                       ])
-    else:
-        report.extend([("deepspeed wheel compiled w.", f"torch {torch_info['version']} ")])
-
-    report.append(("shared memory (/dev/shm) size", get_shm_size()))
-
-    print("DeepSpeed general environment info:")
-    for name, value in report:
-        warns = []
-        if isinstance(value, tuple):
-            value, warns = value
-        print(name, "." * (max_dots - len(name)), value)
-        if warns:
-            for warn in warns:
-                print(warn)
-
-
-def parse_arguments():
-    parser = argparse.ArgumentParser()
-    parser.add_argument('--hide_operator_status',
-                        action='store_true',
-                        help='Suppress display of installation and compatibility statuses of DeepSpeed operators. ')
-    parser.add_argument('--hide_errors_and_warnings', action='store_true', help='Suppress warning and error messages.')
-    args = parser.parse_args()
-    return args
-
-
-def main(hide_operator_status=False, hide_errors_and_warnings=False):
-    if not hide_operator_status:
-        op_report(verbose=not hide_errors_and_warnings)
-    debug_report()
-
-
-def cli_main():
-    args = parse_arguments()
-    main(hide_operator_status=args.hide_operator_status, hide_errors_and_warnings=args.hide_errors_and_warnings)
-
-
-if __name__ == "__main__":
-    main()

Deepspeed-Windows/DeepSpeed readme.txt

@@ -1,18 +0,0 @@
-These are the modified files you would need for a MANUAL build of DeepSpeed 11.1 or 11.2.
-
-They were created by a mix of instructions from erew123 and S95Sedan.
-
-
-- You still have to download the DeepSpeed v11.1 or v11.2 installation files from Microsoft to use these files.
-
-https://github.com/microsoft/DeepSpeed/releases/tag/v0.11.2
-
-
-- These files were provided by S95Sedan here: 
-
-https://github.com/S95Sedan/Deepspeed-Windows
-
-
-- They are related to the manual build of DeepSpeed for AllTalk, which is linked here: 
-
-https://github.com/erew123/alltalk_tts/tree/main?tab=readme-ov-file#-option-2---a-bit-more-complicated

Dockerfile

@@ -0,0 +1,12 @@
+FROM ubuntu:22.04
+ENV HOST 0.0.0.0
+RUN apt-get update && apt-get upgrade -y \
+    && apt-get install -y git build-essential portaudio19-dev \
+    python3 python3-pip gcc wget \
+    && ln -s /usr/bin/python3 /usr/bin/python
+WORKDIR /app
+COPY . .
+RUN pip install --no-cache-dir -r requirements_other.txt
+EXPOSE 7851
+RUN chmod +x launch.sh
+ENTRYPOINT ["sh", "-c", "./launch.sh"]

atsetup.bat

@@ -61,28 +61,36 @@ echo    environment. If you have NOT done this, please run %L_GREEN%cmd_windows.
 echo    in the %L_GREEN%text-generation-webui%RESET% folder and then re-run this script.
 echo.
 echo    BASE REQUIREMENTS
-echo    1) Install the requirements for a %L_GREEN%Nvidia machine%RESET%.
+echo    1) Apply/Re-Apply the requirements for %L_GREEN%Text-generation-webui%RESET%.
-echo    2) Install the requirements for a %L_GREEN%AMD or MAC machine%RESET%.
 echo.
 echo    OPTIONAL
-echo    3) Install the Finetuning requirements.
+echo    2) Git Pull the latest AllTalk updates from Github
-echo    4) Install DeepSpeed v11.2 for CUDA %L_GREEN%11.8%RESET% and Python-3.11.x.
+echo.
-echo    5) Install DeepSpeed v11.2 for CUDA %L_GREEN%12.1%RESET% and Python-3.11.x.
+echo    DEEPSPEED FOR %L_YELLOW=%PyTorch 2.1.x%RESET%
-echo    6) Uninstall DeepSpeed.
+echo    4) Install DeepSpeed v11.2 for CUDA %L_GREEN%11.8%RESET% and Python-3.11.x and %L_YELLOW%PyTorch 2.1.x%RESET%.
+echo    5) Install DeepSpeed v11.2 for CUDA %L_GREEN%12.1%RESET% and Python-3.11.x and %L_YELLOW%PyTorch 2.1.x%RESET%.
+echo.
+echo    DEEPSPEED FOR %L_YELLOW=%PyTorch 2.2.x%RESET% (March 2024 builds of Text-gen-webui and later)
+echo    6) Install DeepSpeed v14.0 for CUDA %L_GREEN%12.1%RESET% and Python-3.11.x and %L_YELLOW%PyTorch 2.2.x%RESET%.
+echo    7) Install DeepSpeed v14.0 for CUDA %L_GREEN%11.8%RESET% and Python-3.11.x and %L_YELLOW%PyTorch 2.2.x%RESET%.
+echo.
+echo    U) Uninstall DeepSpeed.
 echo.
 echo.   OTHER
-echo    7) Generate a diagnostics file.
+echo    8) Generate a diagnostics file.
 echo.
 echo    9) %L_RED%Exit/Quit%RESET%
 echo.
 set /p WebUIOption="Enter your choice: "
 if "%WebUIOption%"=="1" goto InstallNvidiaTextGen
-if "%WebUIOption%"=="2" goto InstallOtherTextGen
+if "%WebUIOption%"=="2" goto TGGitpull
-if "%WebUIOption%"=="3" goto InstallFinetuneTextGen
 if "%WebUIOption%"=="4" goto InstallDeepSpeed118TextGen
 if "%WebUIOption%"=="5" goto InstallDeepSpeed121TextGen
-if "%WebUIOption%"=="6" goto UnInstallDeepSpeed
+if "%WebUIOption%"=="6" goto InstallDeepSpeed121TextGenPytorch221
-if "%WebUIOption%"=="7" goto GenerateDiagsTextGen
+if "%WebUIOption%"=="7" goto InstallDeepSpeed118TextGenPytorch221
+if "%WebUIOption%"=="U" goto UnInstallDeepSpeed
+if "%WebUIOption%"=="u" goto UnInstallDeepSpeed
+if "%WebUIOption%"=="8" goto GenerateDiagsTextGen
 if "%WebUIOption%"=="9" goto End
 goto WebUIMenu
 
@@ -95,27 +103,31 @@ echo    BASE REQUIREMENTS
 echo    1) Install AllTalk as a Standalone Application
 echo.
 echo    OPTIONAL
-echo    2) Delete AllTalk's custom Python environment
+echo    2) Git Pull the latest AllTalk updates from Github
-echo    3) Install the Finetuning requirements
+echo    3) Re-Apply/Update the requirements file
+echo    4) Delete AllTalk's custom Python environment
+echo    5) Purge the PIP cache
 echo.
 echo.   OTHER
-echo    4) Generate a diagnostics file
+echo    8) Generate a diagnostics file
 echo.
 echo    9) %L_RED%Exit/Quit%RESET%
 echo.
 set /p StandaloneOption="Enter your choice: "
 if "%StandaloneOption%"=="1" goto InstallCustomStandalone
-if "%StandaloneOption%"=="2" goto DeleteCustomStandalone
+if "%StandaloneOption%"=="2" goto STGitpull
-if "%StandaloneOption%"=="3" goto InstallFinetuneStandalone
+if "%StandaloneOption%"=="3" goto STReapplyrequirements
-if "%StandaloneOption%"=="4" goto GenerateDiagsStandalone
+if "%StandaloneOption%"=="4" goto STDeleteCustomStandalone
+if "%StandaloneOption%"=="5" goto STPurgepipcache
+if "%StandaloneOption%"=="8" goto GenerateDiagsStandalone
 if "%StandaloneOption%"=="9" goto End
 goto StandaloneMenu
 
 :InstallNvidiaTextGen
-pip install -r requirements_nvidia.txt
+pip install -r system\requirements\requirements_textgen.txt
 if %ERRORLEVEL% neq 0 (
     echo.
-    echo    There was an error installing the Nvidia requirements.
+    echo    There was an error installing the requirements.
     echo    Press any key to return to the menu.
     echo.
     pause
@@ -123,42 +135,24 @@ if %ERRORLEVEL% neq 0 (
 )
 Echo.
 echo.
-Echo    Nvidia machine requirements installed successfully.
+Echo    Requirements installed successfully.
 Echo. 
 pause
 goto WebUIMenu
 
-:InstallOtherTextGen
+:TGGitpull
-pip install -r requirements_other.txt
+git pull
 if %ERRORLEVEL% neq 0 (
     echo.
-    echo    There was an error installing the Other requirements.
+    echo    There was an error pulling from Github.
     echo    Press any key to return to the menu.
     echo.
     pause
     goto WebUIMenu
 )
 Echo.
-echo.
+echo     AllTalk Updated from Github. Please re-apply
-Echo    Other machine requirements installed successfully.
+echo     the latest requirements file. Option 1
-Echo. 
-pause
-goto WebUIMenu
-
-:InstallFinetuneTextGen
-pip install -r requirements_finetune.txt
-if %ERRORLEVEL% neq 0 (
-    echo.
-    echo    There was an error installing the Finetune requirements.
-    echo    Press any key to return to the menu.
-    echo.
-    pause
-    goto WebUIMenu
-)
-Echo.
-Echo    Finetune requirements installed successfully. Please ensure you have installed CUDA 11.8.
-Echo    Finetune needs access to %L_GREEN%cublas64_11.dll%RESET% and will error if this is not setup correctly.
-Echo    Please see here for details %L_GREEN%https://github.com/erew123/alltalk_tts#-finetuning-a-model%RESET%
 Echo. 
 pause
 goto WebUIMenu
@@ -227,6 +221,70 @@ del deepspeed-0.11.2+cuda121-cp311-cp311-win_amd64.whl
 pause
 goto WebUIMenu
 
+:InstallDeepSpeed121TextGenPytorch221
+echo Downloading DeepSpeed...
+curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-14.0/deepspeed-0.14.0+ce78a63-cp311-cp311-win_amd64.whl
+if %ERRORLEVEL% neq 0 (
+    echo.
+    echo    Failed to download DeepSpeed wheel file.
+    echo    Please check your internet connection or try again later.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto WebUIMenu
+)
+echo    DeepSpeed wheel file downloaded successfully.
+echo    Installing DeepSpeed...
+pip install deepspeed-0.14.0+ce78a63-cp311-cp311-win_amd64.whl
+if %ERRORLEVEL% neq 0 (
+    echo. 
+    echo    Failed to install DeepSpeed.
+    echo    Please check if the wheel file is compatible with your system.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto WebUIMenu
+)
+Echo.
+echo.
+Echo    DeepSpeed installed successfully.
+Echo. 
+del deepspeed-0.14.0+ce78a63-cp311-cp311-win_amd64.whl
+pause
+goto WebUIMenu
+
+:InstallDeepSpeed118TextGenPytorch221
+echo Downloading DeepSpeed...
+curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-14.0/deepspeed-0.14.0+cu118-cp311-cp311-win_amd64.whl
+if %ERRORLEVEL% neq 0 (
+    echo.
+    echo    Failed to download DeepSpeed wheel file.
+    echo    Please check your internet connection or try again later.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto WebUIMenu
+)
+echo    DeepSpeed wheel file downloaded successfully.
+echo    Installing DeepSpeed...
+pip install deepspeed-0.14.0+cu118-cp311-cp311-win_amd64.whl
+if %ERRORLEVEL% neq 0 (
+    echo. 
+    echo    Failed to install DeepSpeed.
+    echo    Please check if the wheel file is compatible with your system.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto WebUIMenu
+)
+Echo.
+echo.
+Echo    DeepSpeed installed successfully.
+Echo. 
+del deepspeed-0.14.0+cu118-cp311-cp311-win_amd64.whl
+pause
+goto WebUIMenu
+
 :UnInstallDeepSpeed
 pip uninstall deepspeed
 if %ERRORLEVEL% neq 0 (
@@ -320,46 +378,20 @@ if not exist "%INSTALL_ENV_DIR%\python.exe" ( echo. && echo Conda environment is
 @rem activate installer env
 call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" || ( echo. && echo Miniconda hook not found. && goto end )
 
-@rem Ask user for the type of requirements to install
 echo.
-echo    %L_BLUE%Choose the type of requirements to install%RESET%
+echo     Downloading and installing PyTorch. This step can take a long time
+echo     depending on your internet connection and hard drive speed. Please
+echo     be patient.
 echo.
-echo    1. Nvidia graphics card machines
+pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121
-echo    2. Other machines (mac, amd, etc)
+echo Installing other requirements.
 echo.
-set /p UserChoice="Enter your choice (1 or 2): "
+pip install -r system\requirements\requirements_standalone.txt
+curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-12.7/deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
+echo Installing DeepSpeed...
+pip install deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
+del deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
 
-@rem Install requirements based on user choice
-if "%UserChoice%" == "1" (
-    echo.
-    echo Installing Nvidia requirements...
-    echo Downloading Pytorch with CUDA..
-    pip install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --upgrade --force-reinstall --index-url https://download.pytorch.org/whl/cu121
-    pip install numpy==1.24.4
-    pip install typing-extensions>=4.8.0
-    pip install networkx<3.0.0,>=2.5.0
-    pip install fsspec>=2023.6.0
-    pip install soundfile==0.12.1
-    pip install uvicorn==0.24.0.post1
-    pip install TTS==0.21.3
-    pip install fastapi==0.104.1
-    pip install Jinja2==3.1.2
-    pip install requests==2.31.0
-    pip install tqdm==4.66.1
-    pip install importlib-metadata==4.8.1
-    pip install packaging==23.2
-    pip install pydantic==1.10.13
-    pip install sounddevice==0.4.6
-    pip install python-multipart==0.0.6
-    echo Downloading DeepSpeed...
-    curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-12.7/deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
-    echo Installing DeepSpeed...
-    pip install deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
-    del deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
-) else (
-    echo Installing requirements for other machines...
-    pip install -r requirements_other.txt
-)
 
 @rem Create start_environment.bat to run AllTalk environment
 echo @echo off > start_environment.bat
@@ -367,7 +399,6 @@ echo cd /D "%~dp0" >> start_environment.bat
 echo set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda >> start_environment.bat
 echo set INSTALL_ENV_DIR=%cd%\alltalk_environment\env >> start_environment.bat
 echo call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" >> start_environment.bat
-echo exit >> start_environment.bat
 @rem Create start_alltalk.bat to run AllTalk
 echo @echo off > start_alltalk.bat
 echo cd /D "%~dp0" >> start_alltalk.bat
@@ -375,15 +406,22 @@ echo set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda >> start_alltalk.bat
 echo set INSTALL_ENV_DIR=%cd%\alltalk_environment\env >> start_alltalk.bat
 echo call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" >> start_alltalk.bat
 echo call python script.py >> start_alltalk.bat
-echo exit >> start_alltalk.bat
+@rem Create start_finetune.bat to run AllTalk
+echo @echo off > start_finetune.bat
+echo cd /D "%~dp0" >> start_finetune.bat
+echo set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda >> start_finetune.bat
+echo set INSTALL_ENV_DIR=%cd%\alltalk_environment\env >> start_finetune.bat
+echo call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%" >> start_finetune.bat
+echo call python finetune.py >> start_finetune.bat
 Echo.
-Echo    start_alltalk.bat has been created.
+Echo    Run %L_YELLOW%start_alltalk.bat%RESET% to start AllTalk.
-Echo    You can now start AllTalk with %L_YELLOW%start_alltalk.bat%RESET%
+Echo    Run %L_YELLOW%start_finetune.bat%RESET% to start Finetuning.
+Echo    Run %L_YELLOW%start_environment.bat%RESET% to start the AllTalk Python environment.
 Echo.
 pause
 goto StandaloneMenu
 
-:DeleteCustomStandalone
+:STDeleteCustomStandalone
 @rem Check if the alltalk_environment directory exists
 if not exist "%cd%\alltalk_environment\" (
     echo.
@@ -417,47 +455,6 @@ Echo.
 pause
 goto StandaloneMenu
 
-:InstallFinetuneStandalone
-cd /D "%~dp0"
-set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda
-set INSTALL_ENV_DIR=%cd%\alltalk_environment\env
-@rem Check if the Conda environment exists
-if not exist "%INSTALL_ENV_DIR%\python.exe" (
-    echo.
-    echo    The Conda environment at "%INSTALL_ENV_DIR%" does not exist.
-    echo    Please install the environment before proceeding.
-    echo.
-    pause
-    goto StandaloneMenu
-)
-@rem Attempt to activate the Conda environment
-call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%"
-if errorlevel 1 (
-    echo. 
-    echo    Failed to activate the Conda environment.
-    echo    Please check your installation and try again.
-    echo.
-    pause
-    goto StandaloneMenu
-)
-@rem Proceed with installing requirements
-pip install -r requirements_finetune.txt
-if %ERRORLEVEL% neq 0 (
-    echo. 
-    echo    There was an error installing the Finetune requirements.
-    echo    Press any key to return to the menu.
-    echo.
-    pause
-    goto StandaloneMenu
-)
-Echo.
-Echo    Finetune requirements installed successfully. Please ensure you have installed CUDA 11.8.
-Echo    Finetune needs access to %L_GREEN%cublas64_11.dll%RESET% and will error if this is not setup correctly.
-Echo    Please see here for details %L_GREEN%https://github.com/erew123/alltalk_tts#-finetuning-a-model%RESET%
-Echo.
-pause
-goto StandaloneMenu
-
 :GenerateDiagsStandalone
 cd /D "%~dp0"
 set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda
@@ -498,6 +495,144 @@ Echo.
 pause
 goto StandaloneMenu
 
+:STReapplyrequirements
+cd /D "%~dp0"
+set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda
+set INSTALL_ENV_DIR=%cd%\alltalk_environment\env
+@rem Check if the Conda environment exists
+if not exist "%INSTALL_ENV_DIR%\python.exe" (
+    echo.
+    echo    The Conda environment at "%INSTALL_ENV_DIR%" does not exist.
+    echo    Please install the environment before proceeding.
+    echo. 
+    pause
+    goto StandaloneMenu
+)
+@rem Attempt to activate the Conda environment
+call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%"
+if errorlevel 1 (
+    echo. 
+    echo    Failed to activate the Conda environment.
+    echo    Please check your installation and try again.
+    echo.
+    pause
+    goto StandaloneMenu
+)
+@rem Run Reapply requirements
+echo.
+echo     Downloading and installing PyTorch. This step can take a long time
+echo     depending on your internet connection and hard drive speed. Please
+echo     be patient.
+echo.
+pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121
+echo Installing other requirements.
+echo.
+pip install -r system\requirements\requirements_standalone.txt
+curl -LO https://github.com/erew123/alltalk_tts/releases/download/DeepSpeed-12.7/deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
+echo Installing DeepSpeed...
+pip install deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
+del deepspeed-0.12.7+d058d4b-cp311-cp311-win_amd64.whl
+if %ERRORLEVEL% neq 0 (
+    echo.
+    echo    There was an error.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto StandaloneMenu
+)
+Echo.
+Echo.
+Echo    Requirements have been re-applied/updated.
+Echo.
+pause
+goto StandaloneMenu
+
+:STPurgepipcache
+cd /D "%~dp0"
+set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda
+set INSTALL_ENV_DIR=%cd%\alltalk_environment\env
+@rem Check if the Conda environment exists
+if not exist "%INSTALL_ENV_DIR%\python.exe" (
+    echo.
+    echo    The Conda environment at "%INSTALL_ENV_DIR%" does not exist.
+    echo    Please install the environment before proceeding.
+    echo. 
+    pause
+    goto StandaloneMenu
+)
+@rem Attempt to activate the Conda environment
+call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%"
+if errorlevel 1 (
+    echo. 
+    echo    Failed to activate the Conda environment.
+    echo    Please check your installation and try again.
+    echo.
+    pause
+    goto StandaloneMenu
+)
+@rem Clear the PIP cache
+echo.
+echo     Purging the PIP cache of downloaded files.
+echo.
+pip cache purge
+if %ERRORLEVEL% neq 0 (
+    echo.
+    echo    There was an error.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto StandaloneMenu
+)
+Echo.
+Echo    The PIP cache has been purged.
+Echo.
+pause
+goto StandaloneMenu
+
+:STGitpull
+cd /D "%~dp0"
+set CONDA_ROOT_PREFIX=%cd%\alltalk_environment\conda
+set INSTALL_ENV_DIR=%cd%\alltalk_environment\env
+@rem Check if the Conda environment exists
+if not exist "%INSTALL_ENV_DIR%\python.exe" (
+    echo.
+    echo    The Conda environment at "%INSTALL_ENV_DIR%" does not exist.
+    echo    Please install the environment before proceeding.
+    echo. 
+    pause
+    goto StandaloneMenu
+)
+@rem Attempt to activate the Conda environment
+call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%"
+if errorlevel 1 (
+    echo. 
+    echo    Failed to activate the Conda environment.
+    echo    Please check your installation and try again.
+    echo.
+    pause
+    goto StandaloneMenu
+)
+@rem Pull from Github
+echo.
+echo     Pulling the latest updates. Please re-apply
+echo     the latest requirements file. Option 3
+echo.
+git pull
+if %ERRORLEVEL% neq 0 (
+    echo.
+    echo    There was an error pulling from Github.
+    echo    Press any key to return to the menu.
+    echo.
+    pause
+    goto StandaloneMenu
+)
+Echo.
+echo     AllTalk Updated from Github. Please re-apply
+echo     the latest requirements file. Option 3
+Echo.
+pause
+goto StandaloneMenu
+
 :End
 echo Exiting AllTalk Setup Utility...
 exit /b

atsetup.sh

@@ -64,11 +64,12 @@ webui_menu() {
         echo -e "    in the ${L_GREEN}text-generation-webui${NC} folder and then re-run this script."
         echo
         echo "    BASE REQUIREMENTS"
-        echo -e "    1) Install the AllTalk requirements for an ${L_GREEN}Nvidia GPU machine.${NC}"
+        echo -e "    1) Apply/Re-Apply the requirements for an ${L_GREEN}Text-generation-webui${NC}"
-        echo -e "    2) Install the AllTalk requirements for an ${L_GREEN}AMD or MAC machine.${NC}"
         echo
         echo "    OPTIONAL"
-        echo "    3) Install the Finetuning requirements."
+        echo "    2) Git Pull the latest AllTalk updates from Github"
+        echo
+        echo "    DEEPSPEED"
         echo "    4) Install DeepSpeed."
         echo "    5) Uninstall DeepSpeed."
         echo
@@ -81,8 +82,7 @@ webui_menu() {
 
         case $webui_option in
             1) install_nvidia_textgen ;;
-            2) install_other_textgen ;;
+            2) tg_gitpull ;;
-            3) install_finetune_textgen ;;
             4) install_deepspeed ;;
             5) uninstall_deepspeed ;;
             6) generate_diagnostics_textgen ;;
@@ -93,52 +93,40 @@ webui_menu() {
 }
 
 install_nvidia_textgen() {
-    local requirements_file="requirements_nvidia.txt"
+    local requirements_file="system/requirements/requirements_textgen.txt"
-    echo "    Installing Finetune requirements from $requirements_file..."
+    echo "    Installing Requirements from $requirements_file..."
     if ! pip install -r "$requirements_file"; then
-        echo "    There was an error installing the Nvidia requirements."
+        echo
+        echo "    There was an error pulling from Github."
         echo "    Please check the output for details."
         read -p "    Press any key to return to the menu. " -n 1
         echo
         return
     fi
-    echo "    Nvidia requirements installed successfully."
+    echo
+    echo "    Requirements installed successfully."
+    echo
+    echo -e "    To install ${L_YELLOW}DeepSpeed${NC} on Linux, there are additional"
+    echo -e "    steps required. Please see the Github or documentation on DeepSeed."
+    echo
     read -p "    Press any key to continue. " -n 1
     echo
 }
 
-install_other_textgen() {
+tg_gitpull() {
-    local requirements_file="requirements_other.txt"
-    echo "    Installing Other machine requirements from $requirements_file..."
-    if ! pip install -r "$requirements_file"; then
-        echo "    There was an error installing the Other requirements."
-        echo "    Please check the output for details."
-        read -p "    Press any key to return to the menu. " -n 1
-        echo
-        return
-    fi
-    echo "    Other machine requirements installed successfully."
-    read -p "    Press any key to continue. " -n 1
     echo
-}
+    if ! git pull; then
-
+        echo
-install_finetune_textgen() {
+        echo "    There was an error installing the requirements."
-    local requirements_file="requirements_finetune.txt"
-    echo "    Installing Finetune requirements from $requirements_file..."
-    if ! pip install -r "$requirements_file"; then
-        echo "    There was an error installing the Finetune requirements."
         echo "    Please check the output for details."
+        echo
         read -p "    Press any key to return to the menu. " -n 1
         echo
         return
     fi
     echo
-    echo "    Finetune base requirements installed successfully."
+    echo "    AllTalk Updated from Github. Please re-apply"
-    echo
+    echo "    the latest requirements file. (Option 1)"
-    echo -e "    Finetuning requires access to ${GREEN}Cublas${NC} from the ${GREEN}Nvidia CUDA Toolkit${NC}, so will need installing."
-    echo
-    echo -e "    Instructions can be found here:"
-    echo -e "    ${GREEN}https://github.com/erew123/alltalk_tts?#-finetuning-a-model${NC}"
     echo
     read -p "    Press any key to continue. " -n 1
     echo
@@ -185,12 +173,16 @@ uninstall_deepspeed() {
     echo "Uninstalling DeepSpeed..."
     pip uninstall -y deepspeed
     if [ $? -ne 0 ]; then
+        echo
         echo "    There was an error uninstalling DeepSpeed."
+        echo
         echo "    Press any key to return to the menu."
         read -n 1
         return
     fi
+    echo
     echo "    DeepSpeed uninstalled successfully."
+    echo
     echo "    Press any key to continue."
     read -n 1
 }
@@ -198,13 +190,17 @@ uninstall_deepspeed() {
 generate_diagnostics_textgen() {
     # Run diagnostics
     if ! python diagnostics.py; then
+        echo
         echo "    There was an error running diagnostics."
+        echo
         read -p "    Press any key to return to the menu. " -n 1
         echo
         return
     fi
-
+    echo
-    echo "    Diagnostics completed successfully."
+    echo "    Diagnostics log file generated successfully."
+    echo "    Please see diagnostics.log"
+    echo
     read -p "    Press any key to continue. " -n 1
     echo
 }
@@ -220,12 +216,16 @@ standalone_menu() {
         echo "    1) Install AllTalk as a Standalone Application"
         echo
         echo "    OPTIONAL"
-        echo "    2) Delete AllTalk's custom Python environment"
+        echo "    2) Git Pull the latest AllTalk updates from Github"
-        echo "    3) Install the Finetuning requirements"
+        echo "    3) Re-Apply/Update the requirements file"
-        echo "    4) Install DeepSpeed."
+        echo "    4) Delete AllTalk's custom Python environment"
+        echo "    5) Purge the PIP cache"
+        echo
+        echo "    DEEPSPEED"
+        echo "    6) DeepSpeed Instructions/Install"
         echo
         echo "    OTHER"        
-        echo "    5) Generate a diagnostics file"
+        echo "    8) Generate a diagnostics file"
         echo
         echo -e "    9)${L_RED} Exit/Quit${NC}"
         echo
@@ -233,16 +233,19 @@ standalone_menu() {
 
         case $standalone_option in
             1) install_custom_standalone ;;
-            2) delete_custom_standalone ;;
+            2) gitpull_standalone ;;
-            3) install_finetune_standalone ;;
+            3) reapply_standalone ;;
-            4) install_deepspeed ;;
+            4) delete_custom_standalone ;;
-            5) generate_diagnostics_standalone ;;
+            5) pippurge_standalone ;;
+            6) install_deepspeed ;;
+            8) generate_diagnostics_standalone ;;
             9) exit 0 ;;
             *) echo "Invalid option"; sleep 2 ;;
         esac
     done
 }
 
+
 install_custom_standalone() {
     cd "$(dirname "${BASH_SOURCE[0]}")"
 
@@ -284,36 +287,15 @@ install_custom_standalone() {
     # Activate the environment and install requirements
     source "${CONDA_ROOT_PREFIX}/etc/profile.d/conda.sh"
     conda activate "${INSTALL_ENV_DIR}"
-
     echo
-    echo -e "    ${L_BLUE}Choose the type of requirements to install${NC}"
+    echo "    Downloading and installing PyTorch. This step can take a long time"
+    echo "    depending on your internet connection and hard drive speed. Please"
+    echo "    be patient."
+    pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121
     echo
-    echo "    1. Nvidia graphics card machines"
+    echo "    Installing additional requirements."
-    echo "    2. Other machines (mac, amd, etc)"
     echo
-    read -p "    Enter your choice (1 or 2): " user_choice
+    pip install -r system/requirements/requirements_standalone.txt
-
-    if [ "$user_choice" == "1" ]; then
-    	pip install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --upgrade --force-reinstall --index-url https://download.pytorch.org/whl/cu121
-    	pip install numpy==1.24.4
-	pip install typing-extensions>=4.8.0
-	pip install networkx<3.0.0,>=2.5.0
-    	pip install fsspec>=2023.6.0
-	pip install soundfile==0.12.1
-    	pip install uvicorn==0.24.0.post1
-    	pip install TTS==0.21.3
-    	pip install fastapi==0.104.1
-    	pip install Jinja2==3.1.2
-    	pip install requests==2.31.0
-    	pip install tqdm==4.66.1
-    	pip install importlib-metadata==4.8.1
-    	pip install packaging==23.2
-    	pip install pydantic==1.10.13
-    	pip install sounddevice==0.4.6
-    	pip install python-multipart==0.0.6
-    else
-        pip install -r requirements_other.txt
-    fi
     # Create start_environment.sh to run AllTalk
     cat << EOF > start_environment.sh
 #!/bin/bash
@@ -330,27 +312,42 @@ unset PYTHONPATH
 unset PYTHONHOME
 export CUDA_PATH="$INSTALL_ENV_DIR"
 export CUDA_HOME="$CUDA_PATH"
+export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'`
 # activate env
 bash --init-file <(echo "source \"$CONDA_ROOT_PREFIX/etc/profile.d/conda.sh\" && conda activate \"$INSTALL_ENV_DIR\"")
 EOF
     # Create start_alltalk.sh to run AllTalk
     cat << EOF > start_alltalk.sh
 #!/bin/bash
+export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'`
 source "${CONDA_ROOT_PREFIX}/etc/profile.d/conda.sh"
 conda activate "${INSTALL_ENV_DIR}"
 python script.py
+EOF
+    # Create start_finetune.sh to run AllTalk
+    cat << EOF > start_finetune.sh
+#!/bin/bash
+export TRAINER_TELEMETRY=0
+export LD_LIBRARY_PATH=`python3 -c 'import os; import nvidia.cublas.lib; import nvidia.cudnn.lib; print(os.path.dirname(nvidia.cublas.lib.__file__) + ":" + os.path.dirname(nvidia.cudnn.lib.__file__))'`
+source "${CONDA_ROOT_PREFIX}/etc/profile.d/conda.sh"
+conda activate "${INSTALL_ENV_DIR}"
+python finetune.py
 EOF
     chmod +x start_alltalk.sh
     chmod +x start_environment.sh
+    chmod +x start_finetune.sh
     echo
     echo
-    echo -e "    start_alltalk.sh has been created."
+    echo -e "    Run ${L_YELLOW}./start_alltalk.sh${NC} to start AllTalk."
-    echo -e "    You can now start AllTalk with ${L_YELLOW}./start_alltalk.sh${NC}"
+    echo -e "    Run ${L_YELLOW}./start_finetune.sh${NC} to start Finetuning."
+    echo -e "    Run ${L_YELLOW}./start_environment.sh${NC} to start the AllTalk Python environment."
+    echo
+    echo -e "    To install ${L_YELLOW}DeepSpeed${NC} on Linux, there are additional"
+    echo -e "    steps required. Please see the Github or documentation on DeepSeed."
     echo
     read -p "    Press any key to continue. " -n 1
 }
 
-
 delete_custom_standalone() {
     local env_dir="$PWD/alltalk_environment"
     # Check if the alltalk_environment directory exists
@@ -377,47 +374,15 @@ delete_custom_standalone() {
     echo
 }
 
-install_finetune_standalone() {
-    local env_dir="$PWD/alltalk_environment"
-    local conda_root_prefix="${env_dir}/conda"
-    local install_env_dir="${env_dir}/env"
-    if [ ! -d "${install_env_dir}" ]; then
-        echo "    The Conda environment at '${install_env_dir}' does not exist."
-        echo "    Please install the environment before proceeding."
-        read -p "    Press any key to return to the menu. " -n 1
-        echo
-        return
-    fi
-    source "${conda_root_prefix}/etc/profile.d/conda.sh"
-    conda activate "${install_env_dir}"
-    local requirements_file="requirements_finetune.txt"
-    echo "Installing Finetune requirements from $requirements_file..."
-    if ! pip install -r "$requirements_file"; then
-        echo "    There was an error installing the Finetune requirements."
-        echo "    Please check the output for details."
-        read -p "    Press any key to return to the menu. " -n 1
-        echo
-        return
-    fi
-    echo
-    echo "    Finetune base requirements installed successfully."
-    echo
-    echo -e "    Finetuning requires access to ${GREEN}Cublas${NC} from the ${GREEN}Nvidia CUDA Toolkit${NC}, so will need installing."
-    echo
-    echo -e "    Instructions can be found here:"
-    echo -e "    ${GREEN}https://github.com/erew123/alltalk_tts?#-finetuning-a-model${NC}"
-    echo
-    read -p "    Press any key to continue. " -n 1
-    echo
-}
-
 generate_diagnostics_standalone() {
     local env_dir="$PWD/alltalk_environment"
     local conda_root_prefix="${env_dir}/conda"
     local install_env_dir="${env_dir}/env"
     if [ ! -d "${install_env_dir}" ]; then
+        echo
         echo "    The Conda environment at '${install_env_dir}' does not exist."
         echo "    Please install the environment before proceeding."
+        echo
         read -p "    Press any key to return to the menu. " -n 1
         echo
         return
@@ -425,16 +390,97 @@ generate_diagnostics_standalone() {
     source "${conda_root_prefix}/etc/profile.d/conda.sh"
     conda activate "${install_env_dir}"
     if ! python diagnostics.py; then
+        echo
         echo "    There was an error running diagnostics."
+        echo
         read -p "    Press any key to return to the menu. " -n 1
         echo
         return
     fi
+    echo
     echo "    Diagnostics completed successfully."
     read -p "    Press any key to continue. " -n 1
     echo
 }
 
+gitpull_standalone() {
+    local env_dir="$PWD/alltalk_environment"
+    local conda_root_prefix="${env_dir}/conda"
+    local install_env_dir="${env_dir}/env"
+    if [ ! -d "${install_env_dir}" ]; then
+        echo
+        echo "    The Conda environment at '${install_env_dir}' does not exist."
+        echo "    Please install the environment before proceeding."
+        echo
+        read -p "    Press any key to return to the menu. " -n 1
+        echo
+        return
+    fi
+    source "${conda_root_prefix}/etc/profile.d/conda.sh"
+    conda activate "${install_env_dir}"
+    if ! git pull; then
+        echo
+        echo "    There was an error pulling from Github."
+        echo
+        read -p "    Press any key to return to the menu. " -n 1
+        echo
+        return
+    fi
+    echo
+    echo "    AllTalk Updated from Github. Please re-apply."
+    echo "    the latest requirements file. (Option 3)"
+    echo
+    read -p "    Press any key to continue. " -n 1
+    echo
+}
+
+pippurge_standalone() {
+    if ! pip cache purge; then
+        echo
+        echo "    There was an error."
+        echo
+        read -p "    Press any key to return to the menu. " -n 1
+        echo
+        return
+    fi
+    echo
+    echo "    The PIP cache has been purged."
+    echo
+    read -p "    Press any key to continue. " -n 1
+    echo
+}
+
+reapply_standalone() {
+    local env_dir="$PWD/alltalk_environment"
+    local conda_root_prefix="${env_dir}/conda"
+    local install_env_dir="${env_dir}/env"
+    if [ ! -d "${install_env_dir}" ]; then
+        echo
+        echo "    The Conda environment at '${install_env_dir}' does not exist."
+        echo "    Please install the environment before proceeding."
+        echo
+        read -p "    Press any key to return to the menu. " -n 1
+        echo
+        return
+    fi
+    source "${conda_root_prefix}/etc/profile.d/conda.sh"
+    conda activate "${install_env_dir}"
+    echo
+    echo "    Downloading and installing PyTorch. This step can take a long time"
+    echo "    depending on your internet connection and hard drive speed. Please"
+    echo "    be patient."
+    pip install torch>=2.2.1+cu121 torchaudio>=2.2.1+cu121 --extra-index-url https://download.pytorch.org/whl/cu121
+    echo
+    echo "    Installing additional requirements."
+    echo
+    pip install -r system/requirements/requirements_standalone.txt
+    echo
+    echo "    Requirements have been re-applied/updated."
+    echo
+    read -p "    Press any key to continue. " -n 1
+    echo
+}
+
 
 # Start the main menu
 main_menu

confignew.json

@@ -1 +1,25 @@
-{"activate": true, "autoplay": true, "branding": "AllTalk ", "narrator_enabled": false, "deepspeed_activate": false, "delete_output_wavs": "Disabled", "ip_address": "127.0.0.1", "language": "English", "low_vram": false, "local_temperature": "0.7", "local_repetition_penalty": "10.0", "tts_model_loaded": true, "tts_model_name": "tts_models/multilingual/multi-dataset/xtts_v2", "narrator_voice": "female_01.wav", "output_folder_wav": "extensions/alltalk_tts/outputs/", "output_folder_wav_standalone": "outputs/", "port_number": "7851", "remove_trailing_dots": false, "show_text": true, "tts_method_api_local": false, "tts_method_api_tts": false, "tts_method_xtts_local": true, "voice": "female_01.wav"}
+{
+    "activate": true,
+    "autoplay": true,
+    "branding": "AllTalk ",
+    "narrator_enabled": false,
+    "deepspeed_activate": false,
+    "delete_output_wavs": "Disabled",
+    "ip_address": "127.0.0.1",
+    "language": "English",
+    "low_vram": false,
+    "local_temperature": "0.7",
+    "local_repetition_penalty": "10.0",
+    "tts_model_loaded": true,
+    "tts_model_name": "tts_models/multilingual/multi-dataset/xtts_v2",
+    "narrator_voice": "female_01.wav",
+    "output_folder_wav": "extensions/alltalk_tts/outputs/",
+    "output_folder_wav_standalone": "outputs/",
+    "port_number": "7851",
+    "remove_trailing_dots": false,
+    "show_text": true,
+    "tts_method_api_local": false,
+    "tts_method_api_tts": false,
+    "tts_method_xtts_local": true,
+    "voice": "female_01.wav"
+}

diagnostics.py

@@ -8,12 +8,15 @@ try:
     import re
     import sys
     import glob
+    import site
     import textwrap
+    import torch
     import packaging.version
     import packaging.specifiers
     from packaging.specifiers import SpecifierSet
     from packaging.specifiers import SpecifierSet
     from packaging.version import parse as parse_version
+    from pathlib import Path
 except ImportError as e:
     print(f"\033[91mError importing module: {e}\033[0m\n")
     print("\033[94mPlease ensure you started the Text-generation-webUI Python environment with either\033[0m")
@@ -21,6 +24,8 @@ except ImportError as e:
     print("\033[94mfrom the text-generation-webui directory, and then try running the diagnostics again.\033[0m")
     exit(1)
 
+this_dir = Path(__file__).parent
+
 try:
     import psutil
 except ImportError:
@@ -29,24 +34,30 @@ except ImportError:
     import psutil
 
 def get_requirements_file():
-    requirements_files = glob.glob("requirements*.txt")
+    this_dir = Path(__file__).parent  # Assuming 'this_dir' is defined as the script's directory
+    requirements_dir = this_dir / 'system' / 'requirements'
+    requirements_files = list(requirements_dir.glob('requirements*.txt'))  # Using pathlib for globbing
+    
     if not requirements_files:
         print("\033[91mNo requirements files found.\033[0m")
         return None
 
     print("\033[94m\nSelect a requirements file to check against (or press Enter for default 'requirements.txt'):\033[0m\n")
-    for i, file in enumerate(requirements_files, start=1):
+    for i, file_path in enumerate(requirements_files, start=1):
-        print(f"{i}. {file}")
+        print(f"    {i}. {file_path.name}")  # Only print the filename
 
     choice = input("\nEnter the number of your choice: ")
     try:
         choice = int(choice) - 1
-        return requirements_files[choice]
+        return str(requirements_files[choice])  # Return the full path as a string
     except (ValueError, IndexError):
-        return "requirements.txt"
+        return str(requirements_dir / "requirements.txt")  # Return the default path as a string
 
 # Set up logging with filemode='w'
-logging.basicConfig(filename='diagnostics.log', filemode='w', level=logging.INFO)
+logging.basicConfig(filename='diagnostics.log',
+                    filemode='w',
+                    level=logging.INFO,
+                    format='%(message)s')  # Custom format
 
 # Function to get GPU information using subprocess
 def get_gpu_info():
@@ -56,6 +67,22 @@ def get_gpu_info():
     except FileNotFoundError:
         return "NVIDIA GPU information not available"
     
+def get_cpu_info():
+    cpu_info = {
+        'physical_cores': psutil.cpu_count(logical=False),
+        'total_cores': psutil.cpu_count(logical=True),
+        'max_frequency': psutil.cpu_freq().max
+    }
+    return cpu_info
+
+def get_disk_info():
+    disk_info = []
+    partitions = psutil.disk_partitions()
+    for p in partitions:
+        usage = psutil.disk_usage(p.mountpoint)
+        disk_info.append(f" Drive: {p.device} | Total: {usage.total / (1024 ** 3):.2f} GB | Used: {usage.used / (1024 ** 3):.2f} GB | Free: {usage.free / (1024 ** 3):.2f} GB | Type: {p.fstype}")
+    return disk_info
+
 # Function to check if a port is in use
 def is_port_in_use(port):
     for conn in psutil.net_connections():
@@ -73,6 +100,38 @@ def satisfies_wildcard(installed_version, required_version):
         return True
     return False
 
+def test_cuda():
+    cuda_available = torch.cuda.is_available()
+    if cuda_available:
+        try:
+            # Attempt to create a tensor on GPU
+            torch.tensor([1.0, 2.0]).cuda()
+            return "Success - CUDA is available and working."
+        except Exception as e:
+            return f"Fail - CUDA is available but not working. Error: {e}"
+    else:
+        return "Fail - CUDA is not available."
+
+def find_files_in_path_with_wildcard(pattern):
+    # Get the site-packages directory of the current Python environment
+    site_packages_path = site.getsitepackages()
+    found_paths = []
+    # Adjust the sub-directory based on the operating system
+    sub_directory = "nvidia/cublas"
+    if platform.system() == "Linux":
+        sub_directory = os.path.join(sub_directory, "lib")
+    else:
+        sub_directory = os.path.join(sub_directory, "bin")
+    # Iterate over each site-packages directory (there can be more than one)
+    for directory in site_packages_path:
+        # Construct the search directory path
+        search_directory = os.path.join(directory, sub_directory)
+        # Use glob to find all files matching the pattern in this directory
+        for file_path in glob.glob(os.path.join(search_directory, pattern)):
+            if os.path.isfile(file_path):  # Ensure it's a file
+                found_paths.append(file_path)
+    return found_paths
+
 # Function to log and print system information
 def log_system_info():
     # System information
@@ -88,6 +147,9 @@ def log_system_info():
 
     # Torch version
     torch_version = torch.__version__
+    cuda_test_result = test_cuda()
+    cpu_info = get_cpu_info()
+    disk_info = get_disk_info()
 
     # System RAM using psutil
     try:
@@ -123,6 +185,10 @@ def log_system_info():
     search_path = sys.path
     path_env = os.environ.get('PATH', 'N/A')
 
+    # Check for cublas
+    file_name = 'cublas64_11.*' if platform.system() == "Windows" else 'libcublas.so.11*'
+    found_paths = find_files_in_path_with_wildcard(file_name)
+
     # Compare with requirements file
     requirements_file = get_requirements_file()
     if requirements_file:
@@ -146,52 +212,72 @@ def log_system_info():
             logging.info(f"NOTE {requirements_file} not found. Skipping version checks.")
 
     # Log and print information
-    logging.info(f"OS Version: {os_version}")
+    logging.info(f"OPERATING SYSTEM:")
-    logging.info(f"Note: Windows 11 will list as build is 10.x.22xxx")
+    logging.info(f" OS Version: {os_version}")
-    logging.info(f"Torch Version: {torch_version}")
+    logging.info(f" Note: Windows 11 will list as build is 10.x.22xxx")
-    logging.info(f"System RAM: {system_ram}")
+    logging.info(f"\nHARDWARE ENVIRONMENT:")
-    logging.info(f"CUDA_HOME: {cuda_home}")
+    logging.info(f" CPU: Physical Cores: {cpu_info['physical_cores']}, Total Cores: {cpu_info['total_cores']}, Max Frequency: {cpu_info['max_frequency']} MHz")
-    logging.info(f"Port Status: {port_status}")
+    logging.info(f" System RAM: {system_ram}")
-    logging.info(f"Python Version: {platform.python_version()}")
+    logging.info(f"\nGPU INFORMATION:{gpu_info}")
-    logging.info(f"Python Version Info: {python_version_info}")
+    logging.info("DISK INFORMATION:")
-    logging.info(f"Python Executable: {python_executable}")
+    for disk in disk_info:
-    logging.info(f"Python Virtual Environment: {python_virtual_env} (Should be N/A when in Text-generation-webui Conda Python environment)")
+        logging.info(disk)
-    logging.info(f"Conda Environment: {conda_env}")
+    logging.info("\nNETWORK PORT:")
+    logging.info(f" Port Status: {port_status}")
+    logging.info("\nCUDA:")
+    logging.info(f" CUDA Working: {cuda_test_result}")
+    logging.info(f" CUDA_HOME: {cuda_home}")
+    if found_paths:
+        logging.info(f" Cublas64_11 Path: {', '.join(found_paths)}")
+    else:
+        logging.info(f" Cublas64_11 Path: Not found in any search path directories.")
+    logging.info("\nPYTHON & PYTORCH:")
+    logging.info(f" Torch Version: {torch_version}")
+    logging.info(f" Python Version: {platform.python_version()}")
+    logging.info(f" Python Version Info: {python_version_info}")
+    logging.info(f" Python Executable: {python_executable}")
+    logging.info(f" Python Virtual Environment: {python_virtual_env} (Should be N/A when in Text-generation-webui Conda Python environment)")
+    logging.info(f" Conda Environment: {conda_env}")
     logging.info("\nPython Search Path:")
     for path in search_path:
         logging.info(f"  {path}")
-    logging.info("\nOS PATH Environment Variable:")
+    logging.info("\nOS SEARCHPATH ENVIRONMENT:")
     for path in path_env.split(';'):
         logging.info(f"  {path}")
     if required_packages:  # Check if the dictionary is not empty
-        logging.info("\nPackage Versions:")
+        logging.info("\nPACKAGE VERSIONS vs REQUIREMENTS FILE:")
         max_package_length = max(len(package) for package in required_packages.keys())
         for package_name, (operator, required_version) in required_packages.items():
             installed_version = installed_packages.get(package_name, 'Not installed')
-            logging.info(f"{package_name.ljust(max_package_length)}  Required: {operator} {required_version.ljust(12)}  Installed: {installed_version}")
+            logging.info(f" {package_name.ljust(max_package_length)}  Required: {operator} {required_version.ljust(12)}  Installed: {installed_version}")
-    logging.info(f"GPU Information:\n{gpu_info}")
+    logging.info("\nPYTHON PACKAGES:")
-    logging.info("Package Versions:")
     for package, version in package_versions.items():
-        logging.info(f"{package}>= {version}")
+        logging.info(f" {package}>= {version}")
 
     # Print to screen
     print(f"\n\033[94mOS Version:\033[0m \033[92m{os_version}\033[0m")
     print(f"\033[94mOS Ver note:\033[0m \033[92m(Windows 11 will say build is 10.x.22xxx)\033[0m")
-    print(f"\033[94mCUDA_HOME:\033[0m \033[92m{cuda_home}\033[0m")
     print(f"\033[94mSystem RAM:\033[0m \033[92m{system_ram}\033[0m")
-    print(f"\033[94mPort Status:\033[0m \033[92m{port_status}\033[0m")
+    for disk in disk_info:
-    print(f"\033[94mTorch Version:\033[0m \033[92m{torch_version}\033[0m")
+        print(f"\033[94mDisk:\033[0m \033[92m{disk}\033[0m")
+    print(f"\033[94m\nGPU Information:\033[0m {gpu_info}")
+    print(f"\033[94mPort Status:\033[0m \033[92m{port_status}\033[0m\n")
+    if "Fail" in cuda_test_result:
+        print(f"\033[91mCUDA Working:\033[0m \033[91m{cuda_test_result}\033[0m")
+    else:
+        print(f"\033[94mCUDA Working:\033[0m \033[92m{cuda_test_result}\033[0m")
+    print(f"\033[94mCUDA_HOME:\033[0m \033[92m{cuda_home}\033[0m")
+    if found_paths:
+        print(f"\033[94mCublas64_11 Path:\033[0m \033[92m{', '.join(found_paths)}\033[0m")
+    else:
+        print(f"\033[94mCublas64_11 Path:\033[0m \033[91mNot found in any search path directories.\033[0m")    
+    print(f"\033[94m\nTorch Version:\033[0m \033[92m{torch_version}\033[0m")
     print(f"\033[94mPython Version:\033[0m \033[92m{platform.python_version()}\033[0m")
-    print(f"\033[94mPython Version Info:\033[0m \033[92m{python_version_info}\033[0m")
     print(f"\033[94mPython Executable:\033[0m \033[92m{python_executable}\033[0m")
-    print(f"\033[94mPython Virtual Environment:\033[0m \033[92m{python_virtual_env}\033[0m (Should be N/A when in Text-generation-webui Conda Python environment)")
     print(f"\033[94mConda Environment:\033[0m \033[92m{conda_env}\033[0m")
     print(f"\n\033[94mPython Search Path:\033[0m")
     for path in search_path:
         print(f"  {path}")
-    print(f"\n\033[94mOS Search PATH Environment Variable:\033[0m")
-    for path in path_env.split(';'):
-        print(f"  {path}")
     if required_packages:  # Check if the dictionary is not empty
         print("\033[94m\nRequirements file package comparison:\033[0m")
         max_package_length = max(len(package) for package in required_packages.keys())
@@ -214,7 +300,7 @@ def log_system_info():
             color_installed = "\033[92m" if condition_met else "\033[91m"
 
             # Print colored output
-            print(f"{package_name.ljust(max_package_length)}  Required: {color_required}{operator} {required_version.ljust(12)}\033[0m  Installed: {color_installed}{installed_version}\033[0m")
+            print(f"  {package_name.ljust(max_package_length)}  Required: {color_required}{operator} {required_version.ljust(12)}\033[0m  Installed: {color_installed}{installed_version}\033[0m")
 
         print("\nOn Nvidia Graphics cards machines, if your \033[92mInstalled\033[0m version of \033[92mTorch\033[0m and \033[92mTorchaudio\033[0m does")    
         print("not have \033[92m+cu118\033[0m (Cuda 11.8) or \033[92m+cu121\033[0m (Cuda 12.1) listed after them, you do not have CUDA")
@@ -231,9 +317,9 @@ def log_system_info():
         """)
         print(explanation.strip())
     print("")
-    print(f"GPU Information:{gpu_info}")
+    print(f"\033[94mDiagnostic log created:\033[0m \033[92mdiagnostics.log. \033[94mA brief summary of results is displayed above on\033[0m")
-    print(f"\033[94mDiagnostic log created:\033[0m \033[92mdiagnostics.log\033[0m")
+    print(f"\033[94mscreen. Please see the log file for more detail.\033[0m")
-    print(f"\033[94mPlease upload the log file with any support ticket.\033[0m")
+    print(f"\033[94m\nPlease upload the log file with any support ticket.\033[0m")
 
 if __name__ == "__main__":
     log_system_info()

docker-compose-cuda.yml

@@ -0,0 +1,20 @@
+version: '3.8'
+services:
+  alltalk-tts:
+    image: erew123/alltalk_tts:cuda
+    restart: unless-stopped
+    ports:
+      - "7851:7851"
+    volumes:
+      - ./outputs:/app/outputs/
+      - ./models:/app/models/
+      - ./voices:/app/voices/
+      - ./finetune/put-voice-samples-in-here:/app/finetune/put-voice-samples-in-here
+      - ./dockerconfig.json:/app/confignew.json
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: all
+              capabilities: [ gpu ]

docker-compose.yml

@@ -0,0 +1,13 @@
+version: '3.8'
+services:
+  alltalk-tts:
+    image: erew123/alltalk_tts:cpu
+    restart: unless-stopped
+    ports:
+      - "7851:7851"
+    volumes:
+      - ./outputs:/app/outputs/
+      - ./models:/app/models/
+      - ./voices:/app/voices/
+      - ./finetune/put-voice-samples-in-here:/app/finetune/put-voice-samples-in-here
+      - ./dockerconfig.json:/app/confignew.json

dockerconfig.json

@@ -0,0 +1,25 @@
+{
+    "activate": true,
+    "autoplay": true,
+    "branding": "AllTalk ",
+    "narrator_enabled": false,
+    "deepspeed_activate": true,
+    "delete_output_wavs": "Disabled",
+    "ip_address": "127.0.0.1",
+    "language": "English",
+    "low_vram": false,
+    "local_temperature": "0.7",
+    "local_repetition_penalty": "10.0",
+    "tts_model_loaded": true,
+    "tts_model_name": "tts_models/multilingual/multi-dataset/xtts_v2",
+    "narrator_voice": "female_01.wav",
+    "output_folder_wav": "extensions/alltalk_tts/outputs/",
+    "output_folder_wav_standalone": "outputs/",
+    "port_number": "7851",
+    "remove_trailing_dots": false,
+    "show_text": true,
+    "tts_method_api_local": false,
+    "tts_method_api_tts": false,
+    "tts_method_xtts_local": true,
+    "voice": "female_01.wav"
+}

finetune/compact.py

@@ -1,26 +0,0 @@
-import torch
-import os
-MODEL_PATH = "best_model.pth"
-# Check if best_model.pth exists
-if not os.path.exists(MODEL_PATH):
-    print(f"\nCannot find \033[94m{MODEL_PATH}\033[0m in the current folder.")
-    
-    # Check if model.pth exists as an alternative
-    ALTERNATIVE_PATH = "model.pth"
-    if os.path.exists(ALTERNATIVE_PATH):
-        print(f"\nIf your model is currently called \033[93m{ALTERNATIVE_PATH}\033[0m please rename it to \033[94mbest_model.pth\033[0m.")
-    else:
-        print(f"Please put \033[91mbest_model.pth\033[0m in the current folder or rename your model file to \033[94mbest_model.pth\033[0m.")
-    exit(1)
-# Attempt to load the model
-try:
-    checkpoint = torch.load(MODEL_PATH, map_location=torch.device("cpu"))
-except Exception as e:
-    print("Error loading checkpoint:", e)
-    raise
-del checkpoint["optimizer"]
-for key in list(checkpoint["model"].keys()):
-    if "dvae" in key:
-        del checkpoint["model"][key]
-torch.save(checkpoint, "model.pth")
-print(f"\nYour compacted model is created \033[93mmodel.pth\033[0m.")

launch.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+python modeldownload.py
+uvicorn tts_server:app --host 0.0.0.0 --port 7851 --workers 1 --proxy-headers &
+sleep 5
+python script.py

modeldownload.py

@@ -45,18 +45,18 @@ else:
     )
 
 # Read the version specifier from requirements_nvidia.txt
-with open(this_dir / "requirements_nvidia.txt", "r") as req_file:
+with open(this_dir / "system" / "requirements" / "requirements_standalone.txt", "r") as req_file:
     requirements = req_file.readlines()
 
 tts_version_required = None
 for req in requirements:
-    if req.startswith("TTS>="):
+    if req.startswith("TTS=="):
-        tts_version_required = req.strip().split(">=")[1]
+        tts_version_required = req.strip().split("==")[1]
         break
 
 if tts_version_required is None:
     raise ValueError(
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m Could not find TTS version specifier in requirements_nvidia.txt"
+        f"[{params['branding']}Startup] \033[91mWarning\033[0m Could not find TTS version specifier in requirements file"
     )
 
 

nvidia.Dockerfile

@@ -0,0 +1,20 @@
+FROM nvidia/cuda:12.1.1-devel-ubuntu22.04
+ENV HOST 0.0.0.0
+RUN apt-get update && apt-get upgrade -y \
+    && apt-get install -y git build-essential portaudio19-dev \
+    python3 python3-pip gcc wget \
+    ocl-icd-opencl-dev opencl-headers clinfo \
+    libclblast-dev libopenblas-dev \
+    && mkdir -p /etc/OpenCL/vendors && echo "libnvidia-opencl.so.1" > /etc/OpenCL/vendors/nvidia.icd \
+    && ln -s /usr/bin/python3 /usr/bin/python
+WORKDIR /app
+COPY . .
+ENV CUDA_DOCKER_ARCH=all
+
+RUN pip install --no-cache-dir -r requirements_nvidia.txt && \
+    pip install --no-cache-dir -r requirements_finetune.txt && \
+    pip install --no-cache-dir -r requirements_other.txt && \
+    pip install --no-cache-dir deepspeed
+EXPOSE 7851
+RUN chmod +x launch.sh
+ENTRYPOINT ["sh", "-c", "./launch.sh"]

requirements_finetune.txt

@@ -1,7 +0,0 @@
-gradio>=3.50.2
-torch>=2.1.0+cu118
-torchaudio>=2.1.0+cu118
-TTS==0.21.3
-faster-whisper>=0.10.0
-tqdm>=4.66.0
-pandas>=1.5.0

requirements_other.txt

@@ -1,15 +0,0 @@
-numpy>=1.22.0
-soundfile>=0.12.1
-uvicorn>=0.24.0.post1
-TTS>=0.21.3
-torch>=2.1.0
-torchaudio>=2.1.0
-fastapi>=0.104.1
-Jinja2>=3.1.2
-requests>=2.31.0
-tqdm>=4.66.1
-importlib-metadata>=4.8.1
-packaging>=23.2
-pydantic>=1.10.13
-sounddevice>=0.4.6
-python-multipart>=0.0.6

script.py

@@ -16,18 +16,28 @@ import numpy as np
 import soundfile as sf
 import uuid
 
+######################################
+#### ALLTALK ALLOWED STARTUP TIME ####
+######################################
+startup_wait_time = 120
+
+# You can change the above setting to a larger number to allow AllTAlk more time to start up. The default setting is 120 seconds (2 minutes).
+# On some older systems you may need to allow AllTalk more time. So you could set it to 240 (4 minutes) which will give AllTalk more to load.
+
 #################################################################
 #### LOAD PARAMS FROM confignew.json - REQUIRED FOR BRANDING ####
 #################################################################
 # STARTUP VARIABLE - Create "this_dir" variable as the current script directory
 this_dir = Path(__file__).parent.resolve()
 
+
 # load config file in and get settings
 def load_config(file_path):
     with open(file_path, "r") as config_file:
         config = json.load(config_file)
     return config
 
+
 config_file_path = this_dir / "confignew.json"
 # Load the params dictionary from the confignew.json file
 params = load_config(config_file_path)
@@ -48,6 +58,7 @@ except ModuleNotFoundError:
 # IMPORT - Attempt Importing DeepSpeed (required for displaying Deepspeed checkbox in gradio)
 try:
     import deepspeed
+
     deepspeed_installed = True
 except ImportError:
     deepspeed_installed = False
@@ -60,6 +71,7 @@ try:
     from modules.logging_colors import logger
     from modules.ui import create_refresh_button
     from modules.utils import gradio
+
     # This is set to check if the script is being run within text generation webui or as a standalone script. False is running as part of text gen web ui or a gradio interface
     running_in_standalone = False
     output_folder_wav = params["output_folder_wav"]
@@ -73,17 +85,19 @@ except ModuleNotFoundError:
 #### STARTUP VARIABLES ####
 ###########################
 # STARTUP VARIABLE - Import languges file for Gradio to be able to display them in the interface
-with open(this_dir / "languages.json", encoding="utf8") as f:
+with open(this_dir / "system" / "config" / "languages.json", encoding="utf8") as f:
     languages = json.load(f)
 # Create a global lock
 process_lock = threading.Lock()
 # Base setting for a possible FineTuned model existing and the loader being available
 tts_method_xtts_ft = False
 
+
 # Gather the voice files
 def get_available_voices():
     return sorted([voice.name for voice in Path(f"{this_dir}/voices").glob("*.wav")])
 
+
 #########################
 #### LICENSE DISPLAY ####
 #########################
@@ -141,6 +155,7 @@ def check_required_files():
     subprocess.run(["python", str(download_script_path)])
     print(f"[{params['branding']}Startup] All required files are present.")
 
+
 # STARTUP Call Check routine
 check_required_files()
 
@@ -155,7 +170,9 @@ finetuned_model = trained_model_directory.exists()
 # If true, this will add a extra option in the Gradio interface for loading Xttsv2 FT
 if finetuned_model:
     required_files = ["model.pth", "config.json", "vocab.json"]
-    finetuned_model = all((trained_model_directory / file).exists() for file in required_files)
+    finetuned_model = all(
+        (trained_model_directory / file).exists() for file in required_files
+    )
 if finetuned_model:
     print(f"[{params['branding']}Startup] Finetuned model \033[93mDetected\033[0m")
 
@@ -173,6 +190,7 @@ elif params["tts_method_xtts_local"] == True:
 # Set the default for Narrated text without asterisk or quotes to be Narrator
 non_quoted_text_is = True
 
+
 ######################
 #### GRADIO STUFF ####
 ######################
@@ -230,64 +248,80 @@ def signal_handler(sig, frame):
 
 # Attach the signal handler to the SIGINT signal (Ctrl+C)
 signal.signal(signal.SIGINT, signal_handler)
-
+# Check if we're running in docker
-# Start the subprocess
+if os.path.isfile("/.dockerenv"):
-process = subprocess.Popen(["python", script_path])
-
-# Check if the subprocess has started successfully
-if process.poll() is None:
-    print(f"[{params['branding']}Startup] TTS Subprocess starting")
-    print(f"[{params['branding']}Startup]")
     print(
-        f"[{params['branding']}Startup] \033[94mSettings & Documentation:\033[00m",
+        f"[{params['branding']}Startup] \033[94mRunning in Docker. Please wait.\033[0m"
-        f"\033[92mhttp://{params['ip_address']}:{params['port_number']}\033[00m",
     )
-    print(f"[{params['branding']}Startup]")
 else:
-    print(
+    # Start the subprocess
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m TTS Subprocess Webserver failing to start process"
+    process = subprocess.Popen(["python", script_path])
-    )
+    # Check if the subprocess has started successfully
-    print(
+    if process.poll() is None:
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m It could be that you have something on port:",
+        print(f"[{params['branding']}Startup] TTS Subprocess starting")
-        params["port_number"],
+        print(f"[{params['branding']}Startup]")
-    )
-    print(
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m Or you have not started in a Python environement with all the necesssary bits installed"
-    )
-    print(
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m Check you are starting Text-generation-webui with either the start_xxxxx file or the Python environment with cmd_xxxxx file."
-    )
-    print(
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m xxxxx is the type of OS you are on e.g. windows, linux or mac."
-    )
-    print(
-        f"[{params['branding']}Startup] \033[91mWarning\033[0m Alternatively, you could check no other Python processes are running that shouldnt be e.g. Restart your computer is the simple way."
-    )
-    # Cleanly kill off this script, but allow text-generation-webui to keep running, albeit without this alltalk_tts
-    sys.exit(1)
-
-timeout = 120  # Adjust the timeout as needed
-
-# Introduce a delay before starting the check loop
-time.sleep(26)  # Wait 26 secs before checking if the tts_server.py has started up.
-start_time = time.time()
-while time.time() - start_time < timeout:
-    try:
-        response = requests.get(f"{base_url}/ready")
-        if response.status_code == 200:
-            break
-    except requests.RequestException as e:
-        # Print the exception for debugging purposes
         print(
-            f"[{params['branding']}Startup] \033[91mWarning\033[0m TTS Subprocess has NOT started up yet, Will keep trying for 120 seconds maximum. Please wait."
+            f"[{params['branding']}Startup] \033[94m {params['branding']}Settings & Documentation:\033[00m",
+            f"\033[92mhttp://{params['ip_address']}:{params['port_number']}\033[00m",
         )
-    time.sleep(4)
+        print(f"[{params['branding']}Startup]")
-else:
+    else:
-    print(
+        print(
-        f"[{params['branding']}Startup] Startup timed out. Check the server logs for more information."
+            f"[{params['branding']}Startup] \033[91mWarning\033[0m TTS Subprocess Webserver failing to start process"
-    )
+        )
-    # Cleanly kill off this script, but allow text-generation-webui to keep running, albeit without this alltalk_tts
+        print(
-    sys.exit(1)
+            f"[{params['branding']}Startup] \033[91mWarning\033[0m It could be that you have something on port:",
+            params["port_number"],
+        )
+        print(
+            f"[{params['branding']}Startup] \033[91mWarning\033[0m Or you have not started in a Python environement with all the necesssary bits installed"
+        )
+        print(
+            f"[{params['branding']}Startup] \033[91mWarning\033[0m Check you are starting Text-generation-webui with either the start_xxxxx file or the Python environment with cmd_xxxxx file."
+        )
+        print(
+            f"[{params['branding']}Startup] \033[91mWarning\033[0m xxxxx is the type of OS you are on e.g. windows, linux or mac."
+        )
+        print(
+            f"[{params['branding']}Startup] \033[91mWarning\033[0m Alternatively, you could check no other Python processes are running that shouldnt be e.g. Restart your computer is the simple way."
+        )
+        # Cleanly kill off this script, but allow text-generation-webui to keep running, albeit without this alltalk_tts
+        sys.exit(1)
+
+    timeout = startup_wait_time  # Gather timeout setting from startup_wait_time
+
+    # Introduce a delay before starting the check loop
+    time.sleep(26)  # Wait 26 secs before checking if the tts_server.py has started up.
+    start_time = time.time()
+    while time.time() - start_time < timeout:
+        try:
+            response = requests.get(f"{base_url}/ready")
+            if response.status_code == 200:
+                break
+        except requests.RequestException as e:
+            # Print the exception for debugging purposes
+            print(
+                f"[{params['branding']}Startup] \033[91mWarning\033[0m TTS Subprocess has NOT started up yet, Will keep trying for {timeout} seconds maximum. Please wait."
+            )
+        time.sleep(5)
+    else:
+        print(
+            f"\n[{params['branding']}Startup] Startup timed out. Full help available here \033[92mhttps://github.com/erew123/alltalk_tts#-help-with-problems\033[0m"
+        )
+        print(
+            f"[{params['branding']}Startup] On older system you may wish to open and edit \033[94mscript.py\033[0m with a text editor and changing the"
+        )
+        print(
+            f"[{params['branding']}Startup] \033[94mstartup_wait_time = 120\033[0m setting to something like \033[94mstartup_wait_time = 240\033[0m as this will allow"
+        )
+        print(
+            f"[{params['branding']}Startup] AllTalk more time to try load the model into your VRAM. Otherise please visit the Github for"
+        )
+        print(
+            f"[{params['branding']}Startup] a list of other possible troubleshooting options."
+        )
+        # Cleanly kill off this script, but allow text-generation-webui to keep running, albeit without this alltalk_tts
+        sys.exit(1)
 
 
 #####################################
@@ -313,26 +347,26 @@ def send_reload_request(tts_method):
                 params["tts_method_xtts_local"] = False
                 params["tts_method_api_tts"] = True
                 params["deepspeed_activate"] = False
-                audio_path = this_dir / "templates" / "apitts.wav"
+                audio_path = this_dir / "system" / "at_sounds" / "apitts.wav"
                 tts_method_xtts_ft = False
             elif tts_method == "API Local":
                 params["tts_method_api_tts"] = False
                 params["tts_method_xtts_local"] = False
                 params["tts_method_api_local"] = True
                 params["deepspeed_activate"] = False
-                audio_path = this_dir / "templates" / "apilocal.wav"
+                audio_path = this_dir / "system" / "at_sounds" / "apilocal.wav"
                 tts_method_xtts_ft = False
             elif tts_method == "XTTSv2 Local":
                 params["tts_method_api_tts"] = False
                 params["tts_method_api_local"] = False
                 params["tts_method_xtts_local"] = True
-                audio_path = this_dir / "templates" / "xttslocal.wav"
+                audio_path = this_dir / "system" / "at_sounds" / "xttslocal.wav"
                 tts_method_xtts_ft = False
             elif tts_method == "XTTSv2 FT":
                 params["tts_method_api_tts"] = False
                 params["tts_method_api_local"] = False
                 params["tts_method_xtts_local"] = False
-                audio_path = this_dir / "templates" / "xttsfinetuned.wav"
+                audio_path = this_dir / "system" / "at_sounds" / "xttsfinetuned.wav"
                 tts_method_xtts_ft = True
         return f'<audio src="file/{audio_path}" controls autoplay></audio>'
     except requests.exceptions.RequestException as e:
@@ -342,6 +376,7 @@ def send_reload_request(tts_method):
         )
         return {"status": "error", "message": str(e)}
 
+
 ##################
 #### LOW VRAM ####
 ##################
@@ -350,9 +385,9 @@ def send_lowvram_request(low_vram):
     try:
         params["tts_model_loaded"] = False
         if low_vram:
-            audio_path = this_dir / "templates" / "lowvramenabled.wav"
+            audio_path = this_dir / "system" / "at_sounds" / "lowvramenabled.wav"
         else:
-            audio_path = this_dir / "templates" / "lowvramdisabled.wav"
+            audio_path = this_dir / "system" / "at_sounds" / "lowvramdisabled.wav"
         url = f"{base_url}/api/lowvramsetting?new_low_vram_value={low_vram}"
         headers = {"Content-Type": "application/json"}
         response = requests.post(url, headers=headers)
@@ -379,9 +414,9 @@ def send_deepspeed_request(deepspeed_param):
     try:
         params["tts_model_loaded"] = False
         if deepspeed_param:
-            audio_path = this_dir / "templates" / "deepspeedenabled.wav"
+            audio_path = this_dir / "system" / "at_sounds" / "deepspeedenabled.wav"
         else:
-            audio_path = this_dir / "templates" / "deepspeeddisabled.wav"
+            audio_path = this_dir / "system" / "at_sounds" / "deepspeeddisabled.wav"
 
         url = f"{base_url}/api/deepspeed?new_deepspeed_value={deepspeed_param}"
         headers = {"Content-Type": "application/json"}
@@ -456,9 +491,13 @@ def combine(audio_files, output_folder, state):
 
     # Save the combined audio to a file with a specified sample rate
     if "character_menu" in state:
-        output_file_path = os.path.join(output_folder, f'{state["character_menu"]}_{int(time.time())}_combined.wav')
+        output_file_path = os.path.join(
+            output_folder, f'{state["character_menu"]}_{int(time.time())}_combined.wav'
+        )
     else:
-        output_file_path = os.path.join(output_folder, f'TTSOUT_{int(time.time())}_combined.wav')
+        output_file_path = os.path.join(
+            output_folder, f"TTSOUT_{int(time.time())}_combined.wav"
+        )
     sf.write(output_file_path, audio, samplerate=sample_rate)
     # Clean up unnecessary files
     for audio_file in audio_files:
@@ -472,7 +511,7 @@ def combine(audio_files, output_folder, state):
 ################################
 # PREVIEW VOICE - Generate Random Sentence if Voice Preview box is empty
 def random_sentence():
-    with open(this_dir / "harvard_sentences.txt") as f:
+    with open(this_dir / "system" / "config" / "harvard_sentences.txt") as f:
         return random.choice(list(f))
 
 
@@ -489,12 +528,19 @@ def voice_preview(string):
     output_file = Path(params["output_folder_wav"]) / "voice_preview.wav"
     # Generate the audio
     language_code = languages.get(params["language"])
+    temperature = params["local_temperature"]
+    repetition_penalty = params["local_repetition_penalty"]
     # Convert the WindowsPath object to a string before using it in JSON payload
     output_file_str = output_file.as_posix()
     # Lock before making the generate request
     with process_lock:
         generate_response = send_generate_request(
-            string, params["voice"], language_code, output_file_str
+            string,
+            params["voice"],
+            language_code,
+            temperature,
+            repetition_penalty,
+            output_file_str,
         )
     # Check if lock is already acquired
     if process_lock.locked():
@@ -511,12 +557,16 @@ def voice_preview(string):
         return f"[{params['branding']}Server] Audio generation failed. Status: {generate_response.get('status')}"
 
 
+#######################
+#### TEXT CLEANING ####
+#######################
+
 def process_text(text):
     # Normalize HTML encoded quotes
     text = html.unescape(text)
 
     # Replace ellipsis with a single dot
-    text = re.sub(r'\.{3,}', '.', text)
+    text = re.sub(r"\.{3,}", ".", text)
 
     # Pattern to identify combined narrator and character speech
     combined_pattern = r'(\*[^*"]+\*|"[^"*]+")'
@@ -531,22 +581,22 @@ def process_text(text):
     for match in re.finditer(combined_pattern, text):
         # Add the text before the match, if any, as ambiguous
         if start < match.start():
-            ambiguous_text = text[start:match.start()].strip()
+            ambiguous_text = text[start : match.start()].strip()
             if ambiguous_text:
-                ordered_parts.append(('ambiguous', ambiguous_text))
+                ordered_parts.append(("ambiguous", ambiguous_text))
 
         # Add the matched part as either narrator or character
         matched_text = match.group(0)
-        if matched_text.startswith('*') and matched_text.endswith('*'):
+        if matched_text.startswith("*") and matched_text.endswith("*"):
-            ordered_parts.append(('narrator', matched_text.strip('*').strip()))
+            ordered_parts.append(("narrator", matched_text.strip("*").strip()))
         elif matched_text.startswith('"') and matched_text.endswith('"'):
-            ordered_parts.append(('character', matched_text.strip('"').strip()))
+            ordered_parts.append(("character", matched_text.strip('"').strip()))
         else:
             # In case of mixed or improperly formatted parts
-            if '*' in matched_text:
+            if "*" in matched_text:
-                ordered_parts.append(('narrator', matched_text.strip('*').strip('"')))
+                ordered_parts.append(("narrator", matched_text.strip("*").strip('"')))
             else:
-                ordered_parts.append(('character', matched_text.strip('"').strip('*')))
+                ordered_parts.append(("character", matched_text.strip('"').strip("*")))
 
         # Update the start of the next segment
         start = match.end()
@@ -555,10 +605,35 @@ def process_text(text):
     if start < len(text):
         ambiguous_text = text[start:].strip()
         if ambiguous_text:
-            ordered_parts.append(('ambiguous', ambiguous_text))
+            ordered_parts.append(("ambiguous", ambiguous_text))
 
     return ordered_parts
 
+
+########################
+#### IMAGE CLEANING ####
+########################
+
+img_pattern = r'<img[^>]*src\s*=\s*["\'][^"\'>]+["\'][^>]*>'
+
+def extract_and_remove_images(text):
+    """
+    Extracts all image data from the text and removes it for clean TTS processing.
+    Returns the cleaned text and the extracted image data.
+    """
+    img_matches = re.findall(img_pattern, text)
+    img_info = "\n".join(img_matches)  # Store extracted image data
+    cleaned_text = re.sub(img_pattern, '', text)  # Remove images from text
+    return cleaned_text, img_info
+
+def reinsert_images(text, img_info):
+    """
+    Reinserts the previously extracted image data back into the text.
+    """
+    if img_info:  # Check if there are images to reinsert
+        text += f"\n\n{img_info}"
+    return text
+
 #################################
 #### TTS STANDARD GENERATION ####
 #################################
@@ -566,19 +641,21 @@ def process_text(text):
 def output_modifier(string, state):
     if not params["activate"]:
         return string
-    original_string = string
+    img_info = ""
-    #print("ORIGINAL STRING IS:", original_string)
+    cleaned_text, img_info = extract_and_remove_images(string)
-    cleaned_string = before_audio_generation(string, params)
+    # print("Cleaned STRING IS:", cleaned_text)
+    cleaned_string = before_audio_generation(cleaned_text, params)
     if cleaned_string is None:
         return
-    string = cleaned_string
     language_code = languages.get(params["language"])
+    temperature = params["local_temperature"]
+    repetition_penalty = params["local_repetition_penalty"]
     # Create a list to store generated audio paths
     audio_files = []
     if process_lock.acquire(blocking=False):
         try:
             if params["narrator_enabled"]:
-                processed_parts = process_text(original_string)
+                processed_parts = process_text(cleaned_string)
 
                 audio_files_all_paragraphs = []
                 for part_type, part in processed_parts:
@@ -587,29 +664,46 @@ def output_modifier(string, state):
                         continue
 
                     # Determine the voice to use based on the part type
-                    if part_type == 'narrator':
+                    if part_type == "narrator":
                         voice_to_use = params["narrator_voice"]
-                        print(f"[{params['branding']}TTSGen] \033[92mNarrator\033[0m")  # Green
+                        print(
-                    elif part_type == 'character':
+                            f"[{params['branding']}TTSGen] \033[92mNarrator\033[0m"
+                        )  # Green
+                    elif part_type == "character":
                         voice_to_use = params["voice"]
-                        print(f"[{params['branding']}TTSGen] \033[36mCharacter\033[0m")  # Yellow
+                        print(
+                            f"[{params['branding']}TTSGen] \033[36mCharacter\033[0m"
+                        )  # Yellow
                     else:
                         # Handle ambiguous parts based on user preference
-                        voice_to_use = params["voice"] if non_quoted_text_is else params["narrator_voice"]
+                        voice_to_use = (
-                        voice_description = "\033[36mCharacter (Text-not-inside)\033[0m" if non_quoted_text_is else "\033[92mNarrator (Text-not-inside)\033[0m"
+                            params["voice"]
+                            if non_quoted_text_is
+                            else params["narrator_voice"]
+                        )
+                        voice_description = (
+                            "\033[36mCharacter (Text-not-inside)\033[0m"
+                            if non_quoted_text_is
+                            else "\033[92mNarrator (Text-not-inside)\033[0m"
+                        )
                         print(f"[{params['branding']}TTSGen] {voice_description}")
 
                     # Replace multiple exclamation marks, question marks, or other punctuation with a single instance
-                    cleaned_part = re.sub(r'([!?.])\1+', r'\1', part)
+                    cleaned_part = re.sub(r"([!?.])\1+", r"\1", part)
                     # Further clean to remove any other unwanted characters
-                    cleaned_part = re.sub(r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\$]', '', cleaned_part)
+                    cleaned_part = re.sub(r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0100-\u017F\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\u4E00-\u9FFF\u3400-\u4DBF\u20000-\u2A6DF\u3000-\u303F\$]', '', cleaned_part)
                     # Remove all newline characters (single or multiple)
-                    cleaned_part = re.sub(r'\n+', ' ', cleaned_part)
+                    cleaned_part = re.sub(r"\n+", " ", cleaned_part)
 
                     # Generate TTS and output to a file
                     output_filename = get_output_filename(state)
                     generate_response = send_generate_request(
-                        cleaned_part, voice_to_use, language_code, output_filename
+                        cleaned_part,
+                        voice_to_use,
+                        language_code,
+                        temperature,
+                        repetition_penalty,
+                        output_filename,
                     )
                     audio_path = generate_response.get("data", {}).get("audio_path")
                     audio_files_all_paragraphs.append(audio_path)
@@ -620,22 +714,31 @@ def output_modifier(string, state):
                 )
             else:
                 # Decode HTML entities first
-                cleaned_part = html.unescape(original_string)
+                cleaned_part = html.unescape(cleaned_string)
                 # Replace multiple instances of certain punctuation marks with a single instance
-                cleaned_part = re.sub(r'([!?.])\1+', r'\1', cleaned_part)
+                cleaned_part = re.sub(r"([!?.])\1+", r"\1", cleaned_part)
                 # Further clean to remove any other unwanted characters
-                cleaned_part = re.sub(r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\$]', '', cleaned_part)
+                cleaned_part = re.sub(r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0100-\u017F\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\u4E00-\u9FFF\u3400-\u4DBF\u20000-\u2A6DF\u3000-\u303F\$]', '', cleaned_part)
                 # Remove all newline characters (single or multiple)
-                cleaned_part = re.sub(r'\n+', ' ', cleaned_part)
+                cleaned_part = re.sub(r"\n+", " ", cleaned_part)
                 # Process the part and give it a non-character name if being used vai API or standalone.
                 if "character_menu" in state:
-                    output_file = Path(f'{params["output_folder_wav"]}/{state["character_menu"]}_{int(time.time())}.wav')
+                    output_file = Path(
+                        f'{params["output_folder_wav"]}/{state["character_menu"]}_{int(time.time())}.wav'
+                    )
                 else:
-                    output_file = Path(f'{params["output_folder_wav"]}/TTSOUT_{int(time.time())}.wav')
+                    output_file = Path(
+                        f'{params["output_folder_wav"]}/TTSOUT_{int(time.time())}.wav'
+                    )
                 output_file_str = output_file.as_posix()
                 output_file = get_output_filename(state)
                 generate_response = send_generate_request(
-                    cleaned_part, params["voice"], language_code, output_file_str
+                    cleaned_part,
+                    params["voice"],
+                    language_code,
+                    temperature,
+                    repetition_penalty,
+                    output_file_str,
                 )
                 audio_path = generate_response.get("data", {}).get("audio_path")
                 final_output_file = audio_path
@@ -654,42 +757,40 @@ def output_modifier(string, state):
         if audio_path:
             # Handle Gradio and playback
             autoplay = "autoplay" if params["autoplay"] else ""
-            string = (
+            string = (f'<audio src="file/{final_output_file}" controls {autoplay}></audio>')
-                f'<audio src="file/{final_output_file}" controls {autoplay}></audio>'
-            )
-
             if params["show_text"]:
-                string += f"\n\n{original_string}"
+                string += reinsert_images(cleaned_string, img_info)
                 shared.processing_message = "*Is typing...*"
-
             return string
         else:
-            print(
+            print(f"[{params['branding']}Server] \033[91mWarning\033[0m No audio path in the response.")
-                f"[{params['branding']}Server] \033[91mWarning\033[0m No audio path in the response."
-            )
     else:
-        print(
+        print(f"[{params['branding']}Server] \033[91mWarning\033[0m Audio generation failed. Status:", generate_response.get("message"),)
-            f"[{params['branding']}Server] \033[91mWarning\033[0m Audio generation failed. Status:",
-            generate_response.get("message"),
-        )
-
 
 def get_output_filename(state):
     if "character_menu" in state:
-        return Path(f'{params["output_folder_wav"]}/{state["character_menu"]}_{str(uuid.uuid4())[:8]}.wav').as_posix()
+        return Path(
+            f'{params["output_folder_wav"]}/{state["character_menu"]}_{str(uuid.uuid4())[:8]}.wav'
+        ).as_posix()
     else:
-        return Path(f'{params["output_folder_wav"]}/TTSOUT_{str(uuid.uuid4())[:8]}.wav').as_posix()
+        return Path(
+            f'{params["output_folder_wav"]}/TTSOUT_{str(uuid.uuid4())[:8]}.wav'
+        ).as_posix()
 
 
 ###############################################
 #### SEND GENERATION REQUEST TO TTS ENGINE ####
 ###############################################
-def send_generate_request(text, voice, language, output_file):
+def send_generate_request(
+    text, voice, language, temperature, repetition_penalty, output_file
+):
     url = f"{base_url}/api/generate"
     payload = {
         "text": text,
         "voice": voice,
         "language": language,
+        "temperature": temperature,
+        "repetition_penalty": repetition_penalty,
         "output_file": output_file,
     }
     headers = {"Content-Type": "application/json"}
@@ -714,12 +815,14 @@ def state_modifier(state):
     state["stream"] = False
     return state
 
+
 def update_narrator_enabled(value):
     if value == "Enabled":
         params["narrator_enabled"] = True
     elif value == "Disabled":
         params["narrator_enabled"] = False
 
+
 def update_non_quoted_text_is(value):
     global non_quoted_text_is
     if value == "Narrator":
@@ -727,6 +830,7 @@ def update_non_quoted_text_is(value):
     elif value == "Char":
         non_quoted_text_is = True
 
+
 def input_modifier(string, state):
     if not params["activate"]:
         return string
@@ -737,20 +841,31 @@ def input_modifier(string, state):
 
 def ui():
     with gr.Accordion(params["branding"] + " TTS (XTTSv2)"):
+        # Activate alltalk_tts, Enable autoplay, Show text
         with gr.Row():
-            activate = gr.Checkbox(value=params['activate'], label='Enable TTS')
+            activate = gr.Checkbox(value=params["activate"], label="Enable TTS")
-            autoplay = gr.Checkbox(value=params['autoplay'], label='Autoplay TTS')
+            autoplay = gr.Checkbox(value=params["autoplay"], label="Autoplay TTS")
-            show_text = gr.Checkbox(value=params['show_text'], label='Show Text')
+            show_text = gr.Checkbox(value=params["show_text"], label="Show Text")
 
+        # Low vram enable, Deepspeed enable, Remove trailing dots
         with gr.Row():
-            low_vram = gr.Checkbox(value=params['low_vram'], label='Enable Low VRAM Mode')
+            low_vram = gr.Checkbox(
+                value=params["low_vram"], label="Enable Low VRAM Mode"
+            )
             low_vram_play = gr.HTML(visible=False)
-            deepspeed_checkbox = gr.Checkbox(value=params['deepspeed_activate'], label='Enable DeepSpeed', visible=deepspeed_installed)
+            deepspeed_checkbox = gr.Checkbox(
+                value=params["deepspeed_activate"],
+                label="Enable DeepSpeed",
+                visible=deepspeed_installed,
+            )
             deepspeed_checkbox_play = gr.HTML(visible=False)
-            remove_trailing_dots = gr.Checkbox(value=params['remove_trailing_dots'], label='Remove trailing "."')
+            remove_trailing_dots = gr.Checkbox(
+                value=params["remove_trailing_dots"], label='Remove trailing "."'
+            )
 
+        # TTS method, Character voice selection
         with gr.Row():
-            model_loader_choices=["API TTS", "API Local", "XTTSv2 Local"]
+            model_loader_choices = ["API TTS", "API Local", "XTTSv2 Local"]
             if finetuned_model:
                 model_loader_choices.append("XTTSv2 FT")
             tts_radio_buttons = gr.Radio(
@@ -761,32 +876,88 @@ def ui():
             tts_radio_buttons_play = gr.HTML(visible=False)
             with gr.Row():
                 available_voices = get_available_voices()
-                default_voice = params["voice"] # Check if the default voice is in the list of available voices
+                default_voice = params[
+                    "voice"
+                ]  # Check if the default voice is in the list of available voices
 
                 if default_voice not in available_voices:
-                    default_voice = available_voices[ 0 ] # Choose the first available voice as the default
+                    default_voice = available_voices[
+                        0
+                    ]  # Choose the first available voice as the default
                 # Add allow_custom_value=True to the Dropdown
-                voice = gr.Dropdown(available_voices, label="Character Voice", value=default_voice, allow_custom_value=True, )
+                voice = gr.Dropdown(
-                create_refresh_button(voice, lambda: None, lambda: {"choices": get_available_voices(), "value": params["voice"],}, "refresh-button",)
+                    available_voices,
+                    label="Character Voice",
+                    value=default_voice,
+                    allow_custom_value=True,
+                )
+                create_refresh_button(
+                    voice,
+                    lambda: None,
+                    lambda: {
+                        "choices": get_available_voices(),
+                        "value": params["voice"],
+                    },
+                    "refresh-button",
+                )
 
+        # Language, Narrator voice
         with gr.Row():
-            language = gr.Dropdown(languages.keys(), label="Language", value=params["language"])
+            language = gr.Dropdown(
+                languages.keys(), label="Language", value=params["language"]
+            )
             with gr.Row():
-                narrator_voice_gr = gr.Dropdown(get_available_voices(), label="Narrator Voice", allow_custom_value=True, value=params["narrator_voice"])
+                narrator_voice_gr = gr.Dropdown(
-                create_refresh_button(narrator_voice_gr, lambda: None, lambda: {"choices": get_available_voices(), "value": params["narrator_voice"],}, "refresh-button")
+                    get_available_voices(),
+                    label="Narrator Voice",
+                    allow_custom_value=True,
+                    value=params["narrator_voice"],
+                )
+                create_refresh_button(
+                    narrator_voice_gr,
+                    lambda: None,
+                    lambda: {
+                        "choices": get_available_voices(),
+                        "value": params["narrator_voice"],
+                    },
+                    "refresh-button",
+                )
 
-        # Temperature and Repetition Penalty (Not yet parsed to api/implemented)
+        # Temperature, Repetition Penalty
-        # with gr.Row():
+        with gr.Row():
-        #     local_temperature_gr = gr.Slider(minimum=0, maximum=1, step=0.05, label="Temperature", value=params["local_temperature"])
+            local_temperature_gr = gr.Slider(
-        #     local_repetition_penalty_gr = gr.Slider(minimum=0, maximum=20, step=0.5, label="Repetition Penalty", value=params["local_repetition_penalty"])
+                minimum=0.05,
+                maximum=1,
+                step=0.05,
+                label="Temperature",
+                value=params["local_temperature"],
+            )
+            local_repetition_penalty_gr = gr.Slider(
+                minimum=0.5,
+                maximum=20,
+                step=0.5,
+                label="Repetition Penalty",
+                value=params["local_repetition_penalty"],
+            )
 
+        # Narrator enable, Non quoted text, Explanation text
         with gr.Row():
             with gr.Row():
-                narrator_enabled_gr = gr.Radio(choices={"Enabled": "true", "Disabled": "false"}, label="Narrator", value="Enabled" if params.get("narrator_enabled") else "Disabled")
+                narrator_enabled_gr = gr.Radio(
-                non_quoted_text_is_gr = gr.Radio(choices={"Character": "true", "Narrator": "false"}, label='Unmarked text NOT inside of * or " is', value="Character" if non_quoted_text_is else "Narrator")
+                    choices={"Enabled": "true", "Disabled": "false"},
+                    label="Narrator",
+                    value="Enabled" if params.get("narrator_enabled") else "Disabled",
+                )
+                non_quoted_text_is_gr = gr.Radio(
+                    choices={"Character": "true", "Narrator": "false"},
+                    label='Unmarked text NOT inside of * or " is',
+                    value="Character" if non_quoted_text_is else "Narrator",
+                )
                 explanation_text = gr.HTML(
-                f"<p>⚙️ <a href='http://{params['ip_address']}:{params['port_number']}'>Settings and Documentation Page</a><a href='http://{params['ip_address']}:{params['port_number']}'></a>⚙️<br>- Low VRAM Mode and Deepspeed take 15 seconds to be enabled or disabled.<br>- The DeepSpeed checkbox is only visible if DeepSpeed is present.</p>")
+                    f"<p>⚙️ <a href='http://{params['ip_address']}:{params['port_number']}'>Settings and Documentation Page</a><a href='http://{params['ip_address']}:{params['port_number']}'></a>⚙️<br>- Low VRAM Mode and Deepspeed take 15 seconds to be enabled or disabled.<br>- The DeepSpeed checkbox is only visible if DeepSpeed is present.</p>"
+                )
 
+        # Preview speech
         with gr.Row():
             preview_text = gr.Text(
                 show_label=False,
@@ -856,23 +1027,34 @@ def ui():
     autoplay.change(lambda x: params.update({"autoplay": x}), autoplay, None)
     low_vram.change(lambda x: params.update({"low_vram": x}), low_vram, None)
     low_vram.change(lambda x: send_lowvram_request(x), low_vram, low_vram_play, None)
-    tts_radio_buttons.change(send_reload_request, tts_radio_buttons, tts_radio_buttons_play, None)
+    tts_radio_buttons.change(
-    deepspeed_checkbox.change(send_deepspeed_request, deepspeed_checkbox, deepspeed_checkbox_play, None)
+        send_reload_request, tts_radio_buttons, tts_radio_buttons_play, None
+    )
+    deepspeed_checkbox.change(
+        send_deepspeed_request, deepspeed_checkbox, deepspeed_checkbox_play, None
+    )
     remove_trailing_dots.change(
         lambda x: params.update({"remove_trailing_dots": x}), remove_trailing_dots, None
     )
     voice.change(lambda x: params.update({"voice": x}), voice, None)
     language.change(lambda x: params.update({"language": x}), language, None)
 
-
+    # TSS Settings
-    # TS Settings (Not yet parsed to api/implemented)
+    local_temperature_gr.change(
-    # local_temperature_gr.change(lambda x: params.update({"local_temperature": x}), local_temperature_gr, None)
+        lambda x: params.update({"local_temperature": x}), local_temperature_gr, None
-    # local_repetition_penalty_gr.change(lambda x: params.update({"local_repetition_penalty": x}), local_repetition_penalty_gr, None)
+    )
+    local_repetition_penalty_gr.change(
+        lambda x: params.update({"local_repetition_penalty": x}),
+        local_repetition_penalty_gr,
+        None,
+    )
 
     # Narrator selection actions
     narrator_enabled_gr.change(update_narrator_enabled, narrator_enabled_gr, None)
     non_quoted_text_is_gr.change(update_non_quoted_text_is, non_quoted_text_is_gr, None)
-    narrator_voice_gr.change(lambda x: params.update({"narrator_voice": x}), narrator_voice_gr, None)
+    narrator_voice_gr.change(
+        lambda x: params.update({"narrator_voice": x}), narrator_voice_gr, None
+    )
 
     # Play preview
     preview_text.submit(voice_preview, preview_text, preview_audio)

system/at_admin/at_settings.html

@@ -89,25 +89,25 @@
     <h3>AllTalk Startup Settings</h3><br>
     <form action="/update-settings" method="post">
 
-        <label for="activate" style="color: #33cc33;">Activate TTS:</label>
+        <label for="activate" style="color: #b273fa;">TG-webui Activate TTS:</label>
         <select id="activate" name="activate" required="">
             <option {% if data.activate %}selected{% endif %} value="true">True</option>
             <option {% if not data.activate %}selected{% endif %} value="false">False</option>
         </select>
 
-        <label for="autoplay" style="color: #33cc33;">&nbsp; &nbsp; Autoplay TTS output:</label>
+        <label for="autoplay" style="color: #b273fa;">&nbsp; &nbsp; TG-webui Autoplay TTS:</label>
         <select id="autoplay" name="autoplay" required="">
             <option {% if data.autoplay %}selected{% endif %} value="true">True</option>
             <option {% if not data.autoplay %}selected{% endif %} value="false">False</option>
         </select>
 
-        <label for="show_text" style="color: #33cc33;"> Show Text on generation:</label>
+        <label for="show_text" style="color: #b273fa;">TG-webui Show Text:</label>
         <select id="show_text" name="show_text" required="">
             <option {% if data.show_text %}selected{% endif %} value="true">True</option>
             <option {% if not data.show_text %}selected{% endif %} value="false">False</option>
         </select>
 
-        <label for="voice" style="color: #33cc33;">&nbsp; &nbsp; Default voice:</label>
+        <label for="voice" style="color: #b273fa;">&nbsp; &nbsp;  TG-webui Default voice:</label>
         <select id="voice" name="voice" required="">
             {% for file in wav_files %}
             <option value="{{ file }}" {% if data.voice==file %}selected{% endif %}>{{ file }}</option>
@@ -115,7 +115,7 @@
         </select>
 
 
-        <label for="language" style="color: #33cc33;">Default Language:</label>
+        <label for="language" style="color: #b273fa;">TG-webui Default Language:</label>
         <select id="language" name="language" required="">
             <option value="Arabic" {% if data.language=='Arabic' %}selected{% endif %}>Arabic</option>
             <option value="Chinese" {% if data.language=='Chinese' %}selected{% endif %}>Chinese</option>
@@ -135,20 +135,20 @@
             <option value="Turkish" {% if data.language=='Turkish' %}selected{% endif %}>Turkish</option>
         </select>
 
-        <label for="narrator_voice" style="color: #33cc33;">&nbsp; &nbsp; Narrator voice:</label>
+        <label for="narrator_voice" style="color: #b273fa;">&nbsp; &nbsp; TG-webui Narrator voice:</label>
         <select id="narrator_voice" name="narrator_voice" required="">
             {% for file in wav_files %}
             <option value="{{ file }}" {% if data.narrator_voice==file %}selected{% endif %}>{{ file }}</option>
             {% endfor %}
         </select>
 
-        <label for="remove_trailing_dots" style="color: #33cc33;"> Remove Trailing Dots:</label>
+        <label for="remove_trailing_dots" style="color: #b273fa;">TG-webui Remove Trailing Dots:</label>
         <select id="remove_trailing_dots" name="remove_trailing_dots" required="">
             <option {% if data.remove_trailing_dots %}selected{% endif %} value="true">True</option>
             <option {% if not data.remove_trailing_dots %}selected{% endif %} value="false">False</option>
         </select>
 
-        <label for="narrator_enabled" style="color: #33cc33;">&nbsp; &nbsp; Narrator enabled:</label>
+        <label for="narrator_enabled" style="color: #b273fa;">&nbsp; &nbsp; TG-webui Narrator enabled:</label>
         <select id="narrator_enabled" name="narrator_enabled" required="">
             <option {% if data.narrator_enabled %}selected{% endif %} value="true">True</option>
             <option {% if not data.narrator_enabled %}selected{% endif %} value="false">False</option>

system/ft_tokenizer/tokenizer.py

@@ -0,0 +1,869 @@
+import os
+import re
+import textwrap
+from functools import cached_property
+
+import pypinyin
+import torch
+from hangul_romanize import Transliter
+from hangul_romanize.rule import academic
+from num2words import num2words
+from spacy.lang.ar import Arabic
+from spacy.lang.en import English
+from spacy.lang.es import Spanish
+from spacy.lang.ja import Japanese
+from spacy.lang.zh import Chinese
+from tokenizers import Tokenizer
+
+from TTS.tts.layers.xtts.zh_num2words import TextNorm as zh_num2words
+
+
+def get_spacy_lang(lang):
+    if lang == "zh":
+        return Chinese()
+    elif lang == "ja":
+        return Japanese()
+    elif lang == "ar":
+        return Arabic()
+    elif lang == "es":
+        return Spanish()
+    else:
+        # For most languages, Enlish does the job
+        return English()
+
+
+def split_sentence(text, lang, text_split_length=250):
+    """Preprocess the input text"""
+    text_splits = []
+    if text_split_length is not None and len(text) >= text_split_length:
+        text_splits.append("")
+        nlp = get_spacy_lang(lang)
+        nlp.add_pipe("sentencizer")
+        doc = nlp(text)
+        for sentence in doc.sents:
+            if len(text_splits[-1]) + len(str(sentence)) <= text_split_length:
+                # if the last sentence + the current sentence is less than the text_split_length
+                # then add the current sentence to the last sentence
+                text_splits[-1] += " " + str(sentence)
+                text_splits[-1] = text_splits[-1].lstrip()
+            elif len(str(sentence)) > text_split_length:
+                # if the current sentence is greater than the text_split_length
+                for line in textwrap.wrap(
+                    str(sentence),
+                    width=text_split_length,
+                    drop_whitespace=True,
+                    break_on_hyphens=False,
+                    tabsize=1,
+                ):
+                    text_splits.append(str(line))
+            else:
+                text_splits.append(str(sentence))
+
+        if len(text_splits) > 1:
+            if text_splits[0] == "":
+                del text_splits[0]
+    else:
+        text_splits = [text.lstrip()]
+
+    return text_splits
+
+
+_whitespace_re = re.compile(r"\s+")
+
+# List of (regular expression, replacement) pairs for abbreviations:
+_abbreviations = {
+    "en": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("mrs", "misess"),
+            ("mr", "mister"),
+            ("dr", "doctor"),
+            ("st", "saint"),
+            ("co", "company"),
+            ("jr", "junior"),
+            ("maj", "major"),
+            ("gen", "general"),
+            ("drs", "doctors"),
+            ("rev", "reverend"),
+            ("lt", "lieutenant"),
+            ("hon", "honorable"),
+            ("sgt", "sergeant"),
+            ("capt", "captain"),
+            ("esq", "esquire"),
+            ("ltd", "limited"),
+            ("col", "colonel"),
+            ("ft", "fort"),
+        ]
+    ],
+    "es": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("sra", "señora"),
+            ("sr", "señor"),
+            ("dr", "doctor"),
+            ("dra", "doctora"),
+            ("st", "santo"),
+            ("co", "compañía"),
+            ("jr", "junior"),
+            ("ltd", "limitada"),
+        ]
+    ],
+    "fr": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("mme", "madame"),
+            ("mr", "monsieur"),
+            ("dr", "docteur"),
+            ("st", "saint"),
+            ("co", "compagnie"),
+            ("jr", "junior"),
+            ("ltd", "limitée"),
+        ]
+    ],
+    "de": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("fr", "frau"),
+            ("dr", "doktor"),
+            ("st", "sankt"),
+            ("co", "firma"),
+            ("jr", "junior"),
+        ]
+    ],
+    "pt": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("sra", "senhora"),
+            ("sr", "senhor"),
+            ("dr", "doutor"),
+            ("dra", "doutora"),
+            ("st", "santo"),
+            ("co", "companhia"),
+            ("jr", "júnior"),
+            ("ltd", "limitada"),
+        ]
+    ],
+    "it": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            # ("sig.ra", "signora"),
+            ("sig", "signore"),
+            ("dr", "dottore"),
+            ("st", "santo"),
+            ("co", "compagnia"),
+            ("jr", "junior"),
+            ("ltd", "limitata"),
+        ]
+    ],
+    "pl": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("p", "pani"),
+            ("m", "pan"),
+            ("dr", "doktor"),
+            ("sw", "święty"),
+            ("jr", "junior"),
+        ]
+    ],
+    "ar": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            # There are not many common abbreviations in Arabic as in English.
+        ]
+    ],
+    "zh": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            # Chinese doesn't typically use abbreviations in the same way as Latin-based scripts.
+        ]
+    ],
+    "cs": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("dr", "doktor"),  # doctor
+            ("ing", "inženýr"),  # engineer
+            ("p", "pan"),  # Could also map to pani for woman but no easy way to do it
+            # Other abbreviations would be specialized and not as common.
+        ]
+    ],
+    "ru": [
+        (re.compile("\\b%s\\b" % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("г-жа", "госпожа"),  # Mrs.
+            ("г-н", "господин"),  # Mr.
+            ("д-р", "доктор"),  # doctor
+            # Other abbreviations are less common or specialized.
+        ]
+    ],
+    "nl": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("dhr", "de heer"),  # Mr.
+            ("mevr", "mevrouw"),  # Mrs.
+            ("dr", "dokter"),  # doctor
+            ("jhr", "jonkheer"),  # young lord or nobleman
+            # Dutch uses more abbreviations, but these are the most common ones.
+        ]
+    ],
+    "tr": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("b", "bay"),  # Mr.
+            ("byk", "büyük"),  # büyük
+            ("dr", "doktor"),  # doctor
+            # Add other Turkish abbreviations here if needed.
+        ]
+    ],
+    "hu": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            ("dr", "doktor"),  # doctor
+            ("b", "bácsi"),  # Mr.
+            ("nőv", "nővér"),  # nurse
+            # Add other Hungarian abbreviations here if needed.
+        ]
+    ],
+    "ko": [
+        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        for x in [
+            # Korean doesn't typically use abbreviations in the same way as Latin-based scripts.
+        ]
+    ],
+    "ja": [
+        (re.compile("\\b%s\\b" % x[0]), x[1])
+        for x in [
+            ("氏", "さん"),  # Mr.
+            ("夫人", "おんなのひと"),  # Mrs.
+            ("博士", "はかせ"),  # Doctor or PhD
+            ("株", "株式会社"),  # Corporation
+            ("有", "有限会社"),  # Limited company
+            ("大学", "だいがく"),   # University
+            ("先生", "せんせい"),   # Teacher/Professor/Master
+            ("君", "くん")   # Used at the end of boys' names to express familiarity or affection.
+        ]
+    ],
+}
+
+
+def expand_abbreviations_multilingual(text, lang="en"):
+    for regex, replacement in _abbreviations[lang]:
+        text = re.sub(regex, replacement, text)
+    return text
+
+
+_symbols_multilingual = {
+    "en": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " and "),
+            ("@", " at "),
+            ("%", " percent "),
+            ("#", " hash "),
+            ("$", " dollar "),
+            ("£", " pound "),
+            ("°", " degree "),
+        ]
+    ],
+    "es": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " y "),
+            ("@", " arroba "),
+            ("%", " por ciento "),
+            ("#", " numeral "),
+            ("$", " dolar "),
+            ("£", " libra "),
+            ("°", " grados "),
+        ]
+    ],
+    "fr": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " et "),
+            ("@", " arobase "),
+            ("%", " pour cent "),
+            ("#", " dièse "),
+            ("$", " dollar "),
+            ("£", " livre "),
+            ("°", " degrés "),
+        ]
+    ],
+    "de": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " und "),
+            ("@", " at "),
+            ("%", " prozent "),
+            ("#", " raute "),
+            ("$", " dollar "),
+            ("£", " pfund "),
+            ("°", " grad "),
+        ]
+    ],
+    "pt": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " e "),
+            ("@", " arroba "),
+            ("%", " por cento "),
+            ("#", " cardinal "),
+            ("$", " dólar "),
+            ("£", " libra "),
+            ("°", " graus "),
+        ]
+    ],
+    "it": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " e "),
+            ("@", " chiocciola "),
+            ("%", " per cento "),
+            ("#", " cancelletto "),
+            ("$", " dollaro "),
+            ("£", " sterlina "),
+            ("°", " gradi "),
+        ]
+    ],
+    "pl": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " i "),
+            ("@", " małpa "),
+            ("%", " procent "),
+            ("#", " krzyżyk "),
+            ("$", " dolar "),
+            ("£", " funt "),
+            ("°", " stopnie "),
+        ]
+    ],
+    "ar": [
+        # Arabic
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " و "),
+            ("@", " على "),
+            ("%", " في المئة "),
+            ("#", " رقم "),
+            ("$", " دولار "),
+            ("£", " جنيه "),
+            ("°", " درجة "),
+        ]
+    ],
+    "zh": [
+        # Chinese
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " 和 "),
+            ("@", " 在 "),
+            ("%", " 百分之 "),
+            ("#", " 号 "),
+            ("$", " 美元 "),
+            ("£", " 英镑 "),
+            ("°", " 度 "),
+        ]
+    ],
+    "cs": [
+        # Czech
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " a "),
+            ("@", " na "),
+            ("%", " procento "),
+            ("#", " křížek "),
+            ("$", " dolar "),
+            ("£", " libra "),
+            ("°", " stupně "),
+        ]
+    ],
+    "ru": [
+        # Russian
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " и "),
+            ("@", " собака "),
+            ("%", " процентов "),
+            ("#", " номер "),
+            ("$", " доллар "),
+            ("£", " фунт "),
+            ("°", " градус "),
+        ]
+    ],
+    "nl": [
+        # Dutch
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " en "),
+            ("@", " bij "),
+            ("%", " procent "),
+            ("#", " hekje "),
+            ("$", " dollar "),
+            ("£", " pond "),
+            ("°", " graden "),
+        ]
+    ],
+    "tr": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " ve "),
+            ("@", " at "),
+            ("%", " yüzde "),
+            ("#", " diyez "),
+            ("$", " dolar "),
+            ("£", " sterlin "),
+            ("°", " derece "),
+        ]
+    ],
+    "hu": [
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " és "),
+            ("@", " kukac "),
+            ("%", " százalék "),
+            ("#", " kettőskereszt "),
+            ("$", " dollár "),
+            ("£", " font "),
+            ("°", " fok "),
+        ]
+    ],
+    "ko": [
+        # Korean
+        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        for x in [
+            ("&", " 그리고 "),
+            ("@", " 에 "),
+            ("%", " 퍼센트 "),
+            ("#", " 번호 "),
+            ("$", " 달러 "),
+            ("£", " 파운드 "),
+            ("°", " 도 "),
+        ]
+    ],
+    "ja": [
+        (re.compile(r"%s" % re.escape(x[0])), x[1])
+        for x in [
+            ("&", " と "),
+            ("@", " アットマーク "),
+            ("%", " パーセント "),
+            ("#", " ナンバー "),
+            ("$", " ドル "),
+            ("£", " ポンド "),
+            ("°", " 度"),
+            ]
+    ],
+}
+
+
+def expand_symbols_multilingual(text, lang="en"):
+    for regex, replacement in _symbols_multilingual[lang]:
+        text = re.sub(regex, replacement, text)
+        text = text.replace("  ", " ")  # Ensure there are no double spaces
+    return text.strip()
+
+
+_ordinal_re = {
+    "en": re.compile(r"([0-9]+)(st|nd|rd|th)"),
+    "es": re.compile(r"([0-9]+)(º|ª|er|o|a|os|as)"),
+    "fr": re.compile(r"([0-9]+)(º|ª|er|re|e|ème)"),
+    "de": re.compile(r"([0-9]+)(st|nd|rd|th|º|ª|\.(?=\s|$))"),
+    "pt": re.compile(r"([0-9]+)(º|ª|o|a|os|as)"),
+    "it": re.compile(r"([0-9]+)(º|°|ª|o|a|i|e)"),
+    "pl": re.compile(r"([0-9]+)(º|ª|st|nd|rd|th)"),
+    "ar": re.compile(r"([0-9]+)(ون|ين|ث|ر|ى)"),
+    "cs": re.compile(r"([0-9]+)\.(?=\s|$)"),  # In Czech, a dot is often used after the number to indicate ordinals.
+    "ru": re.compile(r"([0-9]+)(-й|-я|-е|-ое|-ье|-го)"),
+    "nl": re.compile(r"([0-9]+)(de|ste|e)"),
+    "tr": re.compile(r"([0-9]+)(\.|inci|nci|uncu|üncü|\.)"),
+    "hu": re.compile(r"([0-9]+)(\.|adik|edik|odik|edik|ödik|ödike|ik)"),
+    "ko": re.compile(r"([0-9]+)(번째|번|차|째)"),
+    "ja": re.compile(r"([0-9]+)(番|回|つ|目|等|位)")
+}
+_number_re = re.compile(r"[0-9]+")
+_currency_re = {
+    "USD": re.compile(r"((\$[0-9\.\,]*[0-9]+)|([0-9\.\,]*[0-9]+\$))"),
+    "GBP": re.compile(r"((£[0-9\.\,]*[0-9]+)|([0-9\.\,]*[0-9]+£))"),
+    "EUR": re.compile(r"(([0-9\.\,]*[0-9]+€)|((€[0-9\.\,]*[0-9]+)))"),
+}
+
+_comma_number_re = re.compile(r"\b\d{1,3}(,\d{3})*(\.\d+)?\b")
+_dot_number_re = re.compile(r"\b\d{1,3}(.\d{3})*(\,\d+)?\b")
+_decimal_number_re = re.compile(r"([0-9]+[.,][0-9]+)")
+
+
+def _remove_commas(m):
+    text = m.group(0)
+    if "," in text:
+        text = text.replace(",", "")
+    return text
+
+
+def _remove_dots(m):
+    text = m.group(0)
+    if "." in text:
+        text = text.replace(".", "")
+    return text
+
+
+def _expand_decimal_point(m, lang="en"):
+    amount = m.group(1).replace(",", ".")
+    return num2words(float(amount), lang=lang if lang != "cs" else "cz")
+
+
+def _expand_currency(m, lang="en", currency="USD"):
+    amount = float((re.sub(r"[^\d.]", "", m.group(0).replace(",", "."))))
+    full_amount = num2words(amount, to="currency", currency=currency, lang=lang if lang != "cs" else "cz")
+
+    and_equivalents = {
+        "en": ", ",
+        "es": " con ",
+        "fr": " et ",
+        "de": " und ",
+        "pt": " e ",
+        "it": " e ",
+        "pl": ", ",
+        "cs": ", ",
+        "ru": ", ",
+        "nl": ", ",
+        "ar": ", ",
+        "tr": ", ",
+        "hu": ", ",
+        "ko": ", ",
+    }
+
+    if amount.is_integer():
+        last_and = full_amount.rfind(and_equivalents[lang])
+        if last_and != -1:
+            full_amount = full_amount[:last_and]
+
+    return full_amount
+
+
+def _expand_ordinal(m, lang="en"):
+    return num2words(int(m.group(1)), ordinal=True, lang=lang if lang != "cs" else "cz")
+
+
+def _expand_number(m, lang="en"):
+    return num2words(int(m.group(0)), lang=lang if lang != "cs" else "cz")
+
+
+def expand_numbers_multilingual(text, lang="en"):
+    if lang == "zh":
+        text = zh_num2words()(text)
+    else:
+        if lang in ["en", "ru"]:
+            text = re.sub(_comma_number_re, _remove_commas, text)
+        else:
+            text = re.sub(_dot_number_re, _remove_dots, text)
+        try:
+            text = re.sub(_currency_re["GBP"], lambda m: _expand_currency(m, lang, "GBP"), text)
+            text = re.sub(_currency_re["USD"], lambda m: _expand_currency(m, lang, "USD"), text)
+            text = re.sub(_currency_re["EUR"], lambda m: _expand_currency(m, lang, "EUR"), text)
+        except:
+            pass
+        if lang != "tr":
+            text = re.sub(_decimal_number_re, lambda m: _expand_decimal_point(m, lang), text)
+        text = re.sub(_ordinal_re[lang], lambda m: _expand_ordinal(m, lang), text)
+        text = re.sub(_number_re, lambda m: _expand_number(m, lang), text)
+    return text
+
+
+def lowercase(text):
+    return text.lower()
+
+
+def collapse_whitespace(text):
+    return re.sub(_whitespace_re, " ", text)
+
+
+def multilingual_cleaners(text, lang):
+    text = text.replace('"', "")
+    if lang == "tr":
+        text = text.replace("İ", "i")
+        text = text.replace("Ö", "ö")
+        text = text.replace("Ü", "ü")
+    text = lowercase(text)
+    text = expand_numbers_multilingual(text, lang)
+    text = expand_abbreviations_multilingual(text, lang)
+    text = expand_symbols_multilingual(text, lang=lang)
+    text = collapse_whitespace(text)
+    return text
+
+
+def basic_cleaners(text):
+    """Basic pipeline that lowercases and collapses whitespace without transliteration."""
+    text = lowercase(text)
+    text = collapse_whitespace(text)
+    return text
+
+
+def chinese_transliterate(text):
+    return "".join(
+        [p[0] for p in pypinyin.pinyin(text, style=pypinyin.Style.TONE3, heteronym=False, neutral_tone_with_five=True)]
+    )
+
+
+def japanese_cleaners(text, katsu):
+    text = katsu.romaji(text)
+    text = lowercase(text)
+    return text
+
+
+def korean_transliterate(text):
+    r = Transliter(academic)
+    return r.translit(text)
+
+
+DEFAULT_VOCAB_FILE = os.path.join(os.path.dirname(os.path.realpath(__file__)), "../data/tokenizer.json")
+
+
+class VoiceBpeTokenizer:
+    def __init__(self, vocab_file=None):
+        self.tokenizer = None
+        if vocab_file is not None:
+            self.tokenizer = Tokenizer.from_file(vocab_file)
+        self.char_limits = {
+            "en": 250,
+            "de": 253,
+            "fr": 273,
+            "es": 239,
+            "it": 213,
+            "pt": 203,
+            "pl": 224,
+            "zh": 82,
+            "ar": 166,
+            "cs": 186,
+            "ru": 182,
+            "nl": 251,
+            "tr": 226,
+            "ja": 71,
+            "hu": 224,
+            "ko": 95,
+        }
+
+    @cached_property
+    def katsu(self):
+        import cutlet
+
+        return cutlet.Cutlet()
+
+    def check_input_length(self, txt, lang):
+        lang = lang.split("-")[0]  # remove the region
+        limit = self.char_limits.get(lang, 250)
+        if len(txt) > limit:
+            print(
+                f"[!] Warning: The text length exceeds the character limit of {limit} for language '{lang}', this might cause truncated audio."
+            )
+
+    def preprocess_text(self, txt, lang):
+        if lang in {"ar", "cs", "de", "en", "es", "fr", "hu", "it", "nl", "pl", "pt", "ru", "tr", "zh", "ko"}:
+            txt = multilingual_cleaners(txt, lang)
+            if lang == "zh":
+                txt = chinese_transliterate(txt)
+            if lang == "ko":
+                txt = korean_transliterate(txt)
+        elif lang == "ja":
+            txt = japanese_cleaners(txt, self.katsu)
+        elif lang == "hi":
+            # @manmay will implement this
+            txt = basic_cleaners(txt)
+        else:
+            raise NotImplementedError(f"Language '{lang}' is not supported.")
+        return txt
+
+    def encode(self, txt, lang):
+        lang = lang.split("-")[0]  # remove the region
+        self.check_input_length(txt, lang)
+        txt = self.preprocess_text(txt, lang)
+        lang = "zh-cn" if lang == "zh" else lang
+        txt = f"[{lang}]{txt}"
+        txt = txt.replace(" ", "[SPACE]")
+        return self.tokenizer.encode(txt).ids
+
+    def decode(self, seq):
+        if isinstance(seq, torch.Tensor):
+            seq = seq.cpu().numpy()
+        txt = self.tokenizer.decode(seq, skip_special_tokens=False).replace(" ", "")
+        txt = txt.replace("[SPACE]", " ")
+        txt = txt.replace("[STOP]", "")
+        txt = txt.replace("[UNK]", "")
+        return txt
+
+    def __len__(self):
+        return self.tokenizer.get_vocab_size()
+
+    def get_number_tokens(self):
+        return max(self.tokenizer.get_vocab().values()) + 1
+
+
+def test_expand_numbers_multilingual():
+    test_cases = [
+        # English
+        ("In 12.5 seconds.", "In twelve point five seconds.", "en"),
+        ("There were 50 soldiers.", "There were fifty soldiers.", "en"),
+        ("This is a 1st test", "This is a first test", "en"),
+        ("That will be $20 sir.", "That will be twenty dollars sir.", "en"),
+        ("That will be 20€ sir.", "That will be twenty euro sir.", "en"),
+        ("That will be 20.15€ sir.", "That will be twenty euro, fifteen cents sir.", "en"),
+        ("That's 100,000.5.", "That's one hundred thousand point five.", "en"),
+        # French
+        ("En 12,5 secondes.", "En douze virgule cinq secondes.", "fr"),
+        ("Il y avait 50 soldats.", "Il y avait cinquante soldats.", "fr"),
+        ("Ceci est un 1er test", "Ceci est un premier test", "fr"),
+        ("Cela vous fera $20 monsieur.", "Cela vous fera vingt dollars monsieur.", "fr"),
+        ("Cela vous fera 20€ monsieur.", "Cela vous fera vingt euros monsieur.", "fr"),
+        ("Cela vous fera 20,15€ monsieur.", "Cela vous fera vingt euros et quinze centimes monsieur.", "fr"),
+        ("Ce sera 100.000,5.", "Ce sera cent mille virgule cinq.", "fr"),
+        # German
+        ("In 12,5 Sekunden.", "In zwölf Komma fünf Sekunden.", "de"),
+        ("Es gab 50 Soldaten.", "Es gab fünfzig Soldaten.", "de"),
+        ("Dies ist ein 1. Test", "Dies ist ein erste Test", "de"),  # Issue with gender
+        ("Das macht $20 Herr.", "Das macht zwanzig Dollar Herr.", "de"),
+        ("Das macht 20€ Herr.", "Das macht zwanzig Euro Herr.", "de"),
+        ("Das macht 20,15€ Herr.", "Das macht zwanzig Euro und fünfzehn Cent Herr.", "de"),
+        # Spanish
+        ("En 12,5 segundos.", "En doce punto cinco segundos.", "es"),
+        ("Había 50 soldados.", "Había cincuenta soldados.", "es"),
+        ("Este es un 1er test", "Este es un primero test", "es"),
+        ("Eso le costará $20 señor.", "Eso le costará veinte dólares señor.", "es"),
+        ("Eso le costará 20€ señor.", "Eso le costará veinte euros señor.", "es"),
+        ("Eso le costará 20,15€ señor.", "Eso le costará veinte euros con quince céntimos señor.", "es"),
+        # Italian
+        ("In 12,5 secondi.", "In dodici virgola cinque secondi.", "it"),
+        ("C'erano 50 soldati.", "C'erano cinquanta soldati.", "it"),
+        ("Questo è un 1° test", "Questo è un primo test", "it"),
+        ("Ti costerà $20 signore.", "Ti costerà venti dollari signore.", "it"),
+        ("Ti costerà 20€ signore.", "Ti costerà venti euro signore.", "it"),
+        ("Ti costerà 20,15€ signore.", "Ti costerà venti euro e quindici centesimi signore.", "it"),
+        # Portuguese
+        ("Em 12,5 segundos.", "Em doze vírgula cinco segundos.", "pt"),
+        ("Havia 50 soldados.", "Havia cinquenta soldados.", "pt"),
+        ("Este é um 1º teste", "Este é um primeiro teste", "pt"),
+        ("Isso custará $20 senhor.", "Isso custará vinte dólares senhor.", "pt"),
+        ("Isso custará 20€ senhor.", "Isso custará vinte euros senhor.", "pt"),
+        (
+            "Isso custará 20,15€ senhor.",
+            "Isso custará vinte euros e quinze cêntimos senhor.",
+            "pt",
+        ),  # "cêntimos" should be "centavos" num2words issue
+        # Polish
+        ("W 12,5 sekundy.", "W dwanaście przecinek pięć sekundy.", "pl"),
+        ("Było 50 żołnierzy.", "Było pięćdziesiąt żołnierzy.", "pl"),
+        ("To będzie kosztować 20€ panie.", "To będzie kosztować dwadzieścia euro panie.", "pl"),
+        ("To będzie kosztować 20,15€ panie.", "To będzie kosztować dwadzieścia euro, piętnaście centów panie.", "pl"),
+        # Arabic
+        ("في الـ 12,5 ثانية.", "في الـ اثنا عشر  , خمسون ثانية.", "ar"),
+        ("كان هناك 50 جنديًا.", "كان هناك خمسون جنديًا.", "ar"),
+        # ("ستكون النتيجة $20 يا سيد.", 'ستكون النتيجة عشرون دولار يا سيد.', 'ar'), # $ and € are mising from num2words
+        # ("ستكون النتيجة 20€ يا سيد.", 'ستكون النتيجة عشرون يورو يا سيد.', 'ar'),
+        # Czech
+        ("Za 12,5 vteřiny.", "Za dvanáct celá pět vteřiny.", "cs"),
+        ("Bylo tam 50 vojáků.", "Bylo tam padesát vojáků.", "cs"),
+        ("To bude stát 20€ pane.", "To bude stát dvacet euro pane.", "cs"),
+        ("To bude 20.15€ pane.", "To bude dvacet euro, patnáct centů pane.", "cs"),
+        # Russian
+        ("Через 12.5 секунды.", "Через двенадцать запятая пять секунды.", "ru"),
+        ("Там было 50 солдат.", "Там было пятьдесят солдат.", "ru"),
+        ("Это будет 20.15€ сэр.", "Это будет двадцать евро, пятнадцать центов сэр.", "ru"),
+        ("Это будет стоить 20€ господин.", "Это будет стоить двадцать евро господин.", "ru"),
+        # Dutch
+        ("In 12,5 seconden.", "In twaalf komma vijf seconden.", "nl"),
+        ("Er waren 50 soldaten.", "Er waren vijftig soldaten.", "nl"),
+        ("Dat wordt dan $20 meneer.", "Dat wordt dan twintig dollar meneer.", "nl"),
+        ("Dat wordt dan 20€ meneer.", "Dat wordt dan twintig euro meneer.", "nl"),
+        # Chinese (Simplified)
+        ("在12.5秒内", "在十二点五秒内", "zh"),
+        ("有50名士兵", "有五十名士兵", "zh"),
+        # ("那将是$20先生", '那将是二十美元先生', 'zh'), currency doesn't work
+        # ("那将是20€先生", '那将是二十欧元先生', 'zh'),
+        # Turkish
+        # ("12,5 saniye içinde.", 'On iki virgül beş saniye içinde.', 'tr'), # decimal doesn't work for TR
+        ("50 asker vardı.", "elli asker vardı.", "tr"),
+        ("Bu 1. test", "Bu birinci test", "tr"),
+        # ("Bu 100.000,5.", 'Bu yüz bin virgül beş.', 'tr'),
+        # Hungarian
+        ("12,5 másodperc alatt.", "tizenkettő egész öt tized másodperc alatt.", "hu"),
+        ("50 katona volt.", "ötven katona volt.", "hu"),
+        ("Ez az 1. teszt", "Ez az első teszt", "hu"),
+        # Korean
+        ("12.5 초 안에.", "십이 점 다섯 초 안에.", "ko"),
+        ("50 명의 병사가 있었다.", "오십 명의 병사가 있었다.", "ko"),
+        ("이것은 1 번째 테스트입니다", "이것은 첫 번째 테스트입니다", "ko"),
+    ]
+    for a, b, lang in test_cases:
+        out = expand_numbers_multilingual(a, lang=lang)
+        assert out == b, f"'{out}' vs '{b}'"
+
+
+def test_abbreviations_multilingual():
+    test_cases = [
+        # English
+        ("Hello Mr. Smith.", "Hello mister Smith.", "en"),
+        ("Dr. Jones is here.", "doctor Jones is here.", "en"),
+        # Spanish
+        ("Hola Sr. Garcia.", "Hola señor Garcia.", "es"),
+        ("La Dra. Martinez es muy buena.", "La doctora Martinez es muy buena.", "es"),
+        # French
+        ("Bonjour Mr. Dupond.", "Bonjour monsieur Dupond.", "fr"),
+        ("Mme. Moreau est absente aujourd'hui.", "madame Moreau est absente aujourd'hui.", "fr"),
+        # German
+        ("Frau Dr. Müller ist sehr klug.", "Frau doktor Müller ist sehr klug.", "de"),
+        # Portuguese
+        ("Olá Sr. Silva.", "Olá senhor Silva.", "pt"),
+        ("Dra. Costa, você está disponível?", "doutora Costa, você está disponível?", "pt"),
+        # Italian
+        ("Buongiorno, Sig. Rossi.", "Buongiorno, signore Rossi.", "it"),
+        # ("Sig.ra Bianchi, posso aiutarti?", 'signora Bianchi, posso aiutarti?', 'it'), # Issue with matching that pattern
+        # Polish
+        ("Dzień dobry, P. Kowalski.", "Dzień dobry, pani Kowalski.", "pl"),
+        ("M. Nowak, czy mogę zadać pytanie?", "pan Nowak, czy mogę zadać pytanie?", "pl"),
+        # Czech
+        ("P. Novák", "pan Novák", "cs"),
+        ("Dr. Vojtěch", "doktor Vojtěch", "cs"),
+        # Dutch
+        ("Dhr. Jansen", "de heer Jansen", "nl"),
+        ("Mevr. de Vries", "mevrouw de Vries", "nl"),
+        # Russian
+        ("Здравствуйте Г-н Иванов.", "Здравствуйте господин Иванов.", "ru"),
+        ("Д-р Смирнов здесь, чтобы увидеть вас.", "доктор Смирнов здесь, чтобы увидеть вас.", "ru"),
+        # Turkish
+        ("Merhaba B. Yılmaz.", "Merhaba bay Yılmaz.", "tr"),
+        ("Dr. Ayşe burada.", "doktor Ayşe burada.", "tr"),
+        # Hungarian
+        ("Dr. Szabó itt van.", "doktor Szabó itt van.", "hu"),
+    ]
+
+    for a, b, lang in test_cases:
+        out = expand_abbreviations_multilingual(a, lang=lang)
+        assert out == b, f"'{out}' vs '{b}'"
+
+
+def test_symbols_multilingual():
+    test_cases = [
+        ("I have 14% battery", "I have 14 percent battery", "en"),
+        ("Te veo @ la fiesta", "Te veo arroba la fiesta", "es"),
+        ("J'ai 14° de fièvre", "J'ai 14 degrés de fièvre", "fr"),
+        ("Die Rechnung beträgt £ 20", "Die Rechnung beträgt pfund 20", "de"),
+        ("O meu email é ana&joao@gmail.com", "O meu email é ana e joao arroba gmail.com", "pt"),
+        ("linguaggio di programmazione C#", "linguaggio di programmazione C cancelletto", "it"),
+        ("Moja temperatura to 36.6°", "Moja temperatura to 36.6 stopnie", "pl"),
+        ("Mám 14% baterie", "Mám 14 procento baterie", "cs"),
+        ("Těším se na tebe @ party", "Těším se na tebe na party", "cs"),
+        ("У меня 14% заряда", "У меня 14 процентов заряда", "ru"),
+        ("Я буду @ дома", "Я буду собака дома", "ru"),
+        ("Ik heb 14% batterij", "Ik heb 14 procent batterij", "nl"),
+        ("Ik zie je @ het feest", "Ik zie je bij het feest", "nl"),
+        ("لدي 14% في البطارية", "لدي 14 في المئة في البطارية", "ar"),
+        ("我的电量为 14%", "我的电量为 14 百分之", "zh"),
+        ("Pilim %14 dolu.", "Pilim yüzde 14 dolu.", "tr"),
+        ("Az akkumulátorom töltöttsége 14%", "Az akkumulátorom töltöttsége 14 százalék", "hu"),
+        ("배터리 잔량이 14%입니다.", "배터리 잔량이 14 퍼센트입니다.", "ko"),
+    ]
+
+    for a, b, lang in test_cases:
+        out = expand_symbols_multilingual(a, lang=lang)
+        assert out == b, f"'{out}' vs '{b}'"
+
+
+if __name__ == "__main__":
+    test_expand_numbers_multilingual()
+    test_abbreviations_multilingual()
+    test_symbols_multilingual()

system/requirements/requirements_standalone.txt

@@ -0,0 +1,172 @@
+absl-py==2.1.0
+aiofiles==23.2.1
+aiohttp==3.9.3
+aiosignal==1.3.1
+altair==5.2.0
+annotated-types==0.6.0
+anyascii==0.3.2
+anyio==4.3.0
+attrs==23.2.0
+audioread==3.0.1
+av==11.0.0
+Babel==2.14.0
+bangla==0.0.2
+blinker==1.7.0
+blis==0.7.11
+bnnumerizer==0.0.2
+bnunicodenormalizer==0.1.6
+catalogue==2.0.10
+certifi==2024.2.2
+cffi==1.16.0
+charset-normalizer==3.3.2
+click==8.1.7
+cloudpathlib==0.16.0
+colorama==0.4.6
+coloredlogs==15.0.1
+confection==0.1.4
+contourpy==1.2.0
+coqpit==0.0.17
+ctranslate2==4.1.0
+cutlet==0.4.0
+cycler==0.12.1
+cymem==2.0.8
+Cython==3.0.9
+dateparser==1.1.8
+decorator==5.1.1
+docopt==0.6.2
+einops==0.7.0
+encodec==0.1.1
+fastapi==0.110.0
+faster-whisper==1.0.1
+ffmpy==0.3.2
+filelock==3.13.3
+Flask==3.0.2
+flatbuffers==24.3.25
+fonttools==4.50.0
+frozenlist==1.4.1
+fsspec==2024.3.1
+fugashi==1.3.1
+fuzzywuzzy>=0.18.0
+g2pkk==0.1.2
+gradio==3.50.2
+gradio_client==0.6.1
+grpcio==1.62.1
+gruut==2.2.3
+gruut-ipa==0.13.0
+gruut-lang-de==2.0.0
+gruut-lang-en==2.0.0
+gruut-lang-es==2.0.0
+gruut-lang-fr==2.0.2
+h11==0.14.0
+hangul-romanize==0.1.0
+httpcore==1.0.4
+httpx==0.27.0
+huggingface-hub==0.22.1
+humanfriendly==10.0
+idna==3.6
+importlib_metadata==7.1.0
+importlib_resources==6.4.0
+inflect==7.0.0
+itsdangerous==2.1.2
+jaconv==0.3.4
+jamo==0.4.1
+jieba==0.42.1
+Jinja2==3.1.3
+joblib==1.3.2
+jsonlines==1.2.0
+jsonschema==4.21.1
+jsonschema-specifications==2023.12.1
+kiwisolver==1.4.5
+langcodes==3.3.0
+lazy_loader==0.3
+librosa==0.10.1
+llvmlite==0.42.0
+Markdown==3.6
+MarkupSafe==2.1.5
+matplotlib==3.8.3
+mojimoji==0.0.13
+mpmath==1.3.0
+msgpack==1.0.8
+multidict==6.0.5
+murmurhash==1.0.10
+networkx==2.8.8
+nltk==3.8.1
+num2words==0.5.13
+numba==0.59.1
+numpy==1.26.4
+nvidia-cublas-cu11>=11.11.3.6
+nvidia-cudnn-cu11>=9.0.0.312
+onnxruntime==1.17.1
+orjson==3.9.15
+packaging==24.0
+pandas==1.5.3
+pillow==10.2.0
+platformdirs==4.2.0
+pooch==1.8.1
+preshed==3.0.9
+protobuf==5.26.0
+psutil==5.9.8
+pycparser==2.21
+pydantic==2.6.4
+pydantic_core==2.16.3
+pydub==0.25.1
+pynndescent==0.5.11
+pyparsing==3.1.2
+pypinyin==0.51.0
+pyreadline3==3.4.1
+pysbd==0.3.4
+python-crfsuite==0.9.10
+python-dateutil==2.9.0.post0
+python-Levenshtein>=0.25.0
+python-multipart==0.0.9
+pytz==2024.1
+PyYAML==6.0.1
+referencing==0.34.0
+regex==2023.12.25
+requests==2.31.0
+rpds-py==0.18.0
+safetensors==0.4.2
+scikit-learn==1.4.1.post1
+scipy==1.12.0
+semantic-version==2.10.0
+six==1.16.0
+smart-open==6.4.0
+sniffio==1.3.1
+sounddevice==0.4.6
+soundfile==0.12.1
+soxr==0.3.7
+spacy==3.7.4
+spacy-legacy==3.0.12
+spacy-loggers==1.0.5
+srsly==2.4.8
+starlette==0.36.3
+SudachiDict-core==20240109
+SudachiPy==0.6.8
+sympy==1.12
+tensorboard==2.16.2
+tensorboard-data-server==0.7.2
+thinc==8.2.3
+threadpoolctl==3.4.0
+tokenizers==0.15.2
+toolz==0.12.1
+torch>=2.2.0
+torchaudio>=2.2.0
+tqdm==4.66.2
+trainer==0.0.36
+transformers==4.39.1
+TTS==0.22.0
+typer==0.9.4
+typing_extensions==4.10.0
+tzdata==2024.1
+tzlocal==5.2
+umap-learn==0.5.5
+Unidecode==1.3.8
+unidic-lite==1.0.8
+urllib3==2.2.1
+uvicorn==0.29.0
+wasabi==1.1.2
+weasel==0.3.4
+websockets==11.0.3
+Werkzeug==3.0.1
+yarl==1.9.4
+zipp==3.18.1

system/requirements/requirements_textgen.txt

@@ -1,7 +1,8 @@
 numpy>=1.22.0
 soundfile>=0.12.1
 uvicorn>=0.24.0.post1
-TTS>=0.21.3
+transformers>=4.37.1
+TTS>=0.22.0
 torch>=2.1.0+cu118
 torchaudio>=2.1.0+cu118
 fastapi>=0.104.1
@@ -13,3 +14,11 @@ packaging>=23.2
 pydantic>=1.10.13
 sounddevice>=0.4.6
 python-multipart>=0.0.6
+cutlet>=0.3.0
+unidic-lite>=1.0.8
+faster-whisper>=0.10.0
+pandas>=1.5.0
+nvidia-cublas-cu11>=11.11.3.6
+nvidia-cudnn-cu11>=9.0.0.312
+fuzzywuzzy>=0.18.0
+python-Levenshtein>=0.25.0

system/st_files/alltalk.js

@@ -3,92 +3,33 @@ import { saveTtsProviderSettings } from './index.js';
 
 export { AllTalkTtsProvider };
 
-const css = `
-.settings-row {
-    display: flex;
-    justify-content: space-between; /* Distribute space evenly between the children */
-    align-items: center;
-    width: 100%; /* Full width to spread out the children */
-}
-
-.settings-option {
-    flex: 1; /* Each child will take equal space */
-    margin: 0 10px; /* Adjust spacing as needed */
-}
-
-.endpoint-option {
-    flex: 1; /* Each child will take equal space */
-    margin: 0 10px; /* Adjust spacing as needed */
-    margin-right: 25px;
-    width: 38%; /* Full width to spread out the children */
-}
-
-.website-row {
-    display: flex;
-    justify-content: start; /* Aligns items to the start of the row */
-    align-items: center;
-    margin-top: 10px; /* Top margin */
-    margin-bottom: 10px; /* Bottom margin */
-}
-
-.website-option {
-    flex: 1; /* Allows each option to take equal space */
-    margin-right: 10px; /* Spacing between the two options, adjust as needed */
-    margin-left: 10px; /* Adjust this value to align as desired */
-    /* Remove the individual margin top and bottom declarations if they are set globally on the .settings-row */
-}
-
-.settings-separator {
-    margin-top: 10px;
-    margin-bottom: 10px;
-    padding: 18x;
-    font-weight: bold;
-    border-top: 1px solid #e1e1e1; /* Grey line */
-    border-bottom: 1px solid #e1e1e1; /* Grey line */
-    text-align: center;
-}
-
-.status-message {
-    flex: 1; /* Adjust as needed */
-    margin: 0 10px;
-    /* Additional styling */
-}
-
-.model-endpoint-row {
-    display: flex;
-    justify-content: space-between;
-    align-items: center;
-    width: 100%;
-}
-
-.model-option, .endpoint-option {
-    flex: 1;
-    margin: 0 10px;
-    margin-left: 10px; /* Adjust this value to align as desired */
-    /* Adjust width as needed to fit in one line */
-}
-
-.endpoint-option {
-    /* Adjust width to align with model dropdown */
-    width: 38%;
-}
-
-#status_info {
-    color: lightgreen; /* Or any specific shade of green you prefer */
-}
-`;
-
-const style = document.createElement('style');
-style.type = 'text/css';
-style.appendChild(document.createTextNode(css));
-document.head.appendChild(style);
-
 class AllTalkTtsProvider {
     //########//
     // Config //
     //########//
 
-    settings;
+    settings = {};
+    constructor() {
+        // Initialize with default settings if they are not already set
+        this.settings = {
+            provider_endpoint: this.settings.provider_endpoint || 'http://localhost:7851',
+            language: this.settings.language || 'en',
+            voiceMap: this.settings.voiceMap || {},
+            at_generation_method: this.settings.at_generation_method || 'standard_generation',
+            narrator_enabled: this.settings.narrator_enabled || 'false',
+            at_narrator_text_not_inside: this.settings.at_narrator_text_not_inside || 'narrator',
+            narrator_voice_gen: this.settings.narrator_voice_gen || 'female_01.wav',
+            finetuned_model: this.settings.finetuned_model || 'false'
+        };
+        // Separate property for dynamically updated settings from the server
+        this.dynamicSettings = {
+            modelsAvailable: [],
+            currentModel: '',
+            deepspeed_available: false,
+            deepSpeedEnabled: false,
+            lowVramEnabled: false,
+        };
+    }
     ready = false;
     voices = [];
     separator = '. ';
@@ -114,23 +55,12 @@ class AllTalkTtsProvider {
         'Hindi': 'hi',
     };
 
-    Settings = {
-        provider_endpoint: 'http://localhost:7851',
-        language: 'en',
-        voiceMap: {},
-        at_generation_method: 'standard_generation',
-        narrator_enabled: 'false',
-        at_narrator_text_not_inside: 'narrator',
-        narrator_voice_gen: 'female_01.wav',
-        finetuned_model: 'false'
-    };
-
     get settingsHtml() {
-        let html = `<div class="settings-separator">AllTalk Settings</div>`;
+        let html = `<div class="at-settings-separator">AllTalk Settings</div>`;
 
-        html += `<div class="settings-row">
+        html += `<div class="at-settings-row">
 
-        <div class="settings-option">
+        <div class="at-settings-option">
             <label for="at_generation_method">AllTalk TTS Generation Method</label>
                 <select id="at_generation_method">
                 <option value="standard_generation">Standard Audio Generation (AT Narrator - Optional)</option>
@@ -139,9 +69,9 @@ class AllTalkTtsProvider {
         </div>
         </div>`;
 
-        html += `<div class="settings-row">
+        html += `<div class="at-settings-row">
 
-        <div class="settings-option">
+        <div class="at-settings-option">
             <label for="at_narrator_enabled">AT Narrator</label>
                 <select id="at_narrator_enabled">
                 <option value="true">Enabled</option>
@@ -149,7 +79,7 @@ class AllTalkTtsProvider {
         </select>
         </div>
 
-        <div class="settings-option">
+        <div class="at-settings-option">
             <label for="at_narrator_text_not_inside">Text Not Inside * or " is</label>
                 <select id="at_narrator_text_not_inside">
                 <option value="narrator">Narrator</option>
@@ -159,8 +89,8 @@ class AllTalkTtsProvider {
 
     </div>`;
 
-        html += `<div class="settings-row">
+        html += `<div class="at-settings-row">
-        <div class="settings-option">
+        <div class="at-settings-option">
             <label for="narrator_voice">Narrator Voice</label>
             <select id="narrator_voice">`;
         if (this.voices) {
@@ -170,7 +100,7 @@ class AllTalkTtsProvider {
         }
         html += `</select>
         </div>
-        <div class="settings-option">
+        <div class="at-settings-option">
             <label for="language_options">Language</label>
             <select id="language_options">`;
         for (let language in this.languageLabels) {
@@ -181,8 +111,8 @@ class AllTalkTtsProvider {
     </div>`;
 
 
-        html += `<div class="model-endpoint-row">
+        html += `<div class="at-model-endpoint-row">
-    <div class="model-option">
+    <div class="at-model-option">
             <label for="switch_model">Switch Model</label>
             <select id="switch_model">
             <option value="api_tts">API TTS</option>
@@ -191,49 +121,47 @@ class AllTalkTtsProvider {
         </select>
         </div>
 
-        <div class="endpoint-option">
+        <div class="at-endpoint-option">
             <label for="at_server">AllTalk Endpoint:</label>
-            <input id="at_server" type="text" class="text_pole" maxlength="80" value="${this.Settings.provider_endpoint}"/>
+            <input id="at_server" type="text" class="text_pole" maxlength="80" value="${this.settings.provider_endpoint}"/>
         </div>
    </div>`;
 
 
-        html += `<div class="model-endpoint-row">
+        html += `<div class="at-model-endpoint-row">
-    <div class="settings-option">
+    <div class="at-settings-option">
         <label for="low_vram">Low VRAM</label>
         <input id="low_vram" type="checkbox"/>
     </div>
-    <div class="settings-option">
+    <div class="at-settings-option">
         <label for="deepspeed">DeepSpeed</label>
         <input id="deepspeed" type="checkbox"/>
     </div>
-    <div class="settings-option status-option">
+    <div class="at-settings-option status-option">
         <span>Status: <span id="status_info">Ready</span></span>
     </div>
-    <div class="settings-option empty-option">
+    <div class="at-settings-option empty-option">
         <!-- This div remains empty for spacing -->
     </div>
 </div>`;
 
 
-        html += `<div class="website-row">
+        html += `<div class="at-website-row">
-        <div class="website-option">
+        <div class="at-website-option">
-        <span>AllTalk <a target="_blank" href="${this.Settings.provider_endpoint}">Config & Docs</a>.</span>
+        <span>AllTalk <a target="_blank" href="${this.settings.provider_endpoint}">Config & Docs</a>.</span>
     </div>
 
-    <div class="website-option">
+    <div class="at-website-option">
         <span>AllTalk <a target="_blank" href="https://github.com/erew123/alltalk_tts/">Website</a>.</span>
     </div>
 </div>`;
 
-html += `<div class="website-row">
+        html += `<div class="at-website-row">
-<div class="website-option">
+<div class="at-website-option">
 <span><strong>Text-generation-webui</strong> users - Uncheck <strong>Enable TTS</strong> in Text-generation-webui.</span>
 </div>
 </div>`;
 
-
-
         return html;
     }
 
@@ -244,14 +172,35 @@ html += `<div class="website-row">
 
     async loadSettings(settings) {
         updateStatus('Offline');
-        // Use default settings if no settings are provided
-        this.settings = Object.keys(settings).length === 0 ? this.Settings : settings;
 
+        if (Object.keys(settings).length === 0) {
+            console.info('Using default AllTalk TTS Provider settings');
+        } else {
+            // Populate settings with provided values, ignoring server-provided settings
+            for (const key in settings) {
+                if (key in this.settings) {
+                    this.settings[key] = settings[key];
+                } else {
+                    console.debug(`Ignoring non-user-configurable setting: ${key}`);
+                }
+            }
+        }
+
+        // Update UI elements to reflect the loaded settings
+        $('#at_server').val(this.settings.provider_endpoint);
+        $('#language_options').val(this.settings.language);
+        //$('#voicemap').val(this.settings.voiceMap);
+        $('#at_generation_method').val(this.settings.at_generation_method);
+        $('#at_narrator_enabled').val(this.settings.narrator_enabled);
+        $('#at_narrator_text_not_inside').val(this.settings.at_narrator_text_not_inside);
+        $('#narrator_voice').val(this.settings.narrator_voice_gen);
+
+        console.debug('AllTalkTTS: Settings loaded');
         try {
             // Check if TTS provider is ready
             await this.checkReady();
+            await this.updateSettingsFromServer(); // Fetch dynamic settings from the TTS server
             await this.fetchTtsVoiceObjects(); // Fetch voices only if service is ready
-            await this.updateSettingsFromServer();
             this.updateNarratorVoicesDropdown();
             this.updateLanguageDropdown();
             this.setupEventListeners();
@@ -259,29 +208,23 @@ html += `<div class="website-row">
             updateStatus('Ready');
         } catch (error) {
             console.error("Error loading settings:", error);
-            // Set status to Error if there is an issue in loading settings
             updateStatus('Offline');
         }
     }
 
     applySettingsToHTML() {
-        // Load AllTalk specific settings first
-        const loadedSettings = loadAllTalkSettings();
         // Apply loaded settings or use defaults
-        this.settings = loadedSettings ? { ...this.Settings, ...loadedSettings } : this.Settings;
         const narratorVoiceSelect = document.getElementById('narrator_voice');
         const atNarratorSelect = document.getElementById('at_narrator_enabled');
         const textNotInsideSelect = document.getElementById('at_narrator_text_not_inside');
         const generationMethodSelect = document.getElementById('at_generation_method');
+        this.settings.narrator_voice = this.settings.narrator_voice_gen;
         // Apply settings to Narrator Voice dropdown
         if (narratorVoiceSelect && this.settings.narrator_voice) {
             narratorVoiceSelect.value = this.settings.narrator_voice.replace('.wav', '');
-            this.settings.narrator_voice_gen = this.settings.narrator_voice;
-            //console.log(this.settings.narrator_voice_gen)
         }
         // Apply settings to AT Narrator Enabled dropdown
         if (atNarratorSelect) {
-            //console.log(this.settings.narrator_enabled)
             // Sync the state with the checkbox in index.js
             const ttsPassAsterisksCheckbox = document.getElementById('tts_pass_asterisks'); // Access the checkbox from index.js
             const ttsNarrateQuotedCheckbox = document.getElementById('tts_narrate_quoted'); // Access the checkbox from index.js
@@ -314,18 +257,15 @@ html += `<div class="website-row">
         const languageSelect = document.getElementById('language_options');
         if (languageSelect && this.settings.language) {
             languageSelect.value = this.settings.language;
-            //console.log(this.settings.language)
         }
         // Apply settings to Text Not Inside dropdown
         if (textNotInsideSelect && this.settings.text_not_inside) {
             textNotInsideSelect.value = this.settings.text_not_inside;
             this.settings.at_narrator_text_not_inside = this.settings.text_not_inside;
-            //console.log(this.settings.at_narrator_text_not_inside)
         }
         // Apply settings to Generation Method dropdown
         if (generationMethodSelect && this.settings.at_generation_method) {
             generationMethodSelect.value = this.settings.at_generation_method;
-            //console.log(this.settings.at_generation_method)
         }
         // Additional logic to disable/enable dropdowns based on the selected generation method
         const isStreamingEnabled = this.settings.at_generation_method === 'streaming_enabled';
@@ -347,7 +287,6 @@ html += `<div class="website-row">
             ftOption.textContent = 'XTTSv2 FT';
             modelSelect.appendChild(ftOption);
         }
-        //console.log("Settings applied to HTML.");
     }
 
     //##############################//
@@ -486,7 +425,7 @@ html += `<div class="website-row">
         const languageSelect = document.getElementById('language_options');
         if (languageSelect) {
             // Ensure default language is set
-            this.settings.language = this.Settings.language;
+            this.settings.language = this.settings.language;
 
             languageSelect.innerHTML = '';
             for (let language in this.languageLabels) {
@@ -610,7 +549,7 @@ html += `<div class="website-row">
         if (narratorVoiceSelect) {
             narratorVoiceSelect.addEventListener('change', (event) => {
                 this.settings.narrator_voice_gen = `${event.target.value}.wav`;
-                this.onSettingsChangeAllTalk(); // Save the settings after change
+                this.onSettingsChange(); // Save the settings after change
             });
         }
 
@@ -618,7 +557,7 @@ html += `<div class="website-row">
         if (textNotInsideSelect) {
             textNotInsideSelect.addEventListener('change', (event) => {
                 this.settings.text_not_inside = event.target.value;
-                this.onSettingsChangeAllTalk(); // Save the settings after change
+                this.onSettingsChange(); // Save the settings after change
             });
         }
 
@@ -656,12 +595,11 @@ html += `<div class="website-row">
                     $('#tts_narrate_dialogues').click();
                     $('#tts_narrate_dialogues').trigger('change');
                 }
-                this.onSettingsChangeAllTalk(); // Save the settings after change
+                this.onSettingsChange(); // Save the settings after change
             });
         }
 
 
-
         // Event Listener for AT Generation Method Dropdown
         const atGenerationMethodSelect = document.getElementById('at_generation_method');
         const atNarratorEnabledSelect = document.getElementById('at_narrator_enabled');
@@ -680,7 +618,7 @@ html += `<div class="website-row">
                     atNarratorEnabledSelect.disabled = false;
                 }
                 this.settings.at_generation_method = selectedMethod; // Update the setting here
-                this.onSettingsChangeAllTalk(); // Save the settings after change
+                this.onSettingsChange(); // Save the settings after change
             });
         }
 
@@ -689,9 +627,19 @@ html += `<div class="website-row">
         if (languageSelect) {
             languageSelect.addEventListener('change', (event) => {
                 this.settings.language = event.target.value;
-                this.onSettingsChangeAllTalk(); // Save the settings after change
+                this.onSettingsChange(); // Save the settings after change
             });
         }
+
+        // Listener for AllTalk Endpoint Input
+        const atServerInput = document.getElementById('at_server');
+        if (atServerInput) {
+            atServerInput.addEventListener('input', (event) => {
+                this.settings.provider_endpoint = event.target.value;
+                this.onSettingsChange(); // Save the settings after change
+            });
+        }
+
     }
 
     //#############################//
@@ -699,30 +647,16 @@ html += `<div class="website-row">
     //#############################//
 
     onSettingsChange() {
-        // Update settings that SillyTavern will save
+        // Update settings based on the UI elements
-        this.settings.provider_endpoint = $('#at_server').val();
+        //this.settings.provider_endpoint = $('#at_server').val();
         this.settings.language = $('#language_options').val();
-        saveTtsProviderSettings(); // This function should save settings handled by SillyTavern
+        //this.settings.voiceMap = $('#voicemap').val();
-        // Call the function to handle AllTalk specific settings
+        this.settings.at_generation_method = $('#at_generation_method').val();
-        this.onSettingsChangeAllTalk(); // Save the settings after change
+        this.settings.narrator_enabled = $('#at_narrator_enabled').val();
-    }
+        this.settings.at_narrator_text_not_inside = $('#at_narrator_text_not_inside').val();
-
+        this.settings.narrator_voice_gen = $('#narrator_voice').val();
-    onSettingsChangeAllTalk() {
+        // Save the updated settings
-        // Update AllTalk specific settings and save to localStorage
+        saveTtsProviderSettings();
-        this.settings.narrator_enabled = $('#at_narrator_enabled').val() === 'true';
-        this.settings.narrator_voice = $('#narrator_voice').val() + '.wav'; // assuming you need to append .wav
-        this.settings.text_not_inside = $('#at_narrator_text_not_inside').val(); // "character" or "narrator"
-        this.settings.at_generation_method = $('#at_generation_method').val(); // Streaming or standard
-        this.settings.language = $('#language_options').val(); // Streaming or standard
-
-        // Call the save function with the current settings
-        saveAllTalkSettings({
-            narrator_voice: this.settings.narrator_voice,
-            narrator_enabled: this.settings.narrator_enabled,
-            text_not_inside: this.settings.text_not_inside,
-            at_generation_method: this.settings.at_generation_method,
-            language: this.settings.language
-        });
     }
 
     //#########################//
@@ -802,38 +736,20 @@ html += `<div class="website-row">
     async generateTts(inputText, voiceId) {
         try {
             if (this.settings.at_generation_method === 'streaming_enabled') {
-                // For streaming method
+                // Construct the streaming URL
                 const streamingUrl = `${this.settings.provider_endpoint}/api/tts-generate-streaming?text=${encodeURIComponent(inputText)}&voice=${encodeURIComponent(voiceId)}.wav&language=${encodeURIComponent(this.settings.language)}&output_file=stream_output.wav`;
-                const audioElement = new Audio(streamingUrl);
-                audioElement.play(); // Play the audio stream directly
                 console.log("Streaming URL:", streamingUrl);
-                return new Response(null, {
+        
-                    status: 200,
+                // Return the streaming URL directly
-                    statusText: "OK",
+                return streamingUrl; 
-                    headers: {
-                        "Content-Type": "audio/wav",
-                        "Content-Location": streamingUrl
-                    }
-                });
             } else {
                 // For standard method
                 const outputUrl = await this.fetchTtsGeneration(inputText, voiceId);
-                // Fetch the audio data as a blob from the URL
                 const audioResponse = await fetch(outputUrl);
                 if (!audioResponse.ok) {
                     throw new Error(`HTTP ${audioResponse.status}: Failed to fetch audio data`);
                 }
-                const audioBlob = await audioResponse.blob();
+                return audioResponse; // Return the fetch response directly
-                if (!audioBlob.type.startsWith('audio/')) {
-                    throw new Error(`Invalid audio data format. Expecting audio/*, got ${audioBlob.type}`);
-                }
-                return new Response(audioBlob, {
-                    status: 200,
-                    statusText: "OK",
-                    headers: {
-                        "Content-Type": audioBlob.type
-                    }
-                });
             }
         } catch (error) {
             console.error("Error in generateTts:", error);
@@ -841,6 +757,7 @@ html += `<div class="website-row">
         }
     }
 
+
     //####################//
     //  Generate Standard //
     //####################//
@@ -852,8 +769,8 @@ html += `<div class="website-row">
             'text_filtering': "standard",
             'character_voice_gen': voiceId + ".wav",
             'narrator_enabled': this.settings.narrator_enabled,
-            'narrator_voice_gen': this.settings.narrator_voice_gen,
+            'narrator_voice_gen': this.settings.narrator_voice_gen + ".wav",
-            'text_not_inside': this.settings.text_not_inside,
+            'text_not_inside': this.settings.at_narrator_text_not_inside,
             'language': this.settings.language,
             'output_file_name': "st_output",
             'output_file_timestamp': "true",
@@ -877,7 +794,6 @@ html += `<div class="website-row">
             if (!response.ok) {
                 const errorText = await response.text();
                 console.error(`[fetchTtsGeneration] Error Response Text:`, errorText);
-                // Uncomment the following line if you have a UI element for displaying errors
                 // toastr.error(response.statusText, 'TTS Generation Failed');
                 throw new Error(`HTTP ${response.status}: ${errorText}`);
             }
@@ -912,27 +828,6 @@ function updateStatus(message) {
             case 'Error':
                 statusElement.style.color = 'red';
                 break;
-            // Add more cases as needed
         }
     }
 }
-
-//########################//
-//  Save/load AT Settings //
-//########################//
-
-function saveAllTalkSettings(settingsToSave) {
-    // Save the specific settings to localStorage
-    console.log("settings", settingsToSave)
-    localStorage.setItem('AllTalkSettings', JSON.stringify(settingsToSave));
-}
-
-function loadAllTalkSettings() {
-    // Retrieve the settings from localStorage
-    const settings = localStorage.getItem('AllTalkSettings');
-    // If settings exist, parse them back into an object
-    if (settings) {
-        return JSON.parse(settings);
-    }
-    return null;
-}

system/st_files/index.js

@@ -1,6 +1,6 @@
 import { callPopup, cancelTtsPlay, eventSource, event_types, name2, saveSettingsDebounced } from '../../../script.js';
 import { ModuleWorkerWrapper, doExtrasFetch, extension_settings, getApiUrl, getContext, modules } from '../../extensions.js';
-import { delay, escapeRegex, getStringHash } from '../../utils.js';
+import { delay, escapeRegex, getBase64Async, getStringHash, onlyUnique } from '../../utils.js';
 import { EdgeTtsProvider } from './edge.js';
 import { ElevenLabsTtsProvider } from './elevenlabs.js';
 import { SileroTtsProvider } from './silerotts.js';
@@ -318,12 +318,14 @@ async function playAudioData(audioBlob) {
     if (currentAudioJob == null) {
         console.log('Cancelled TTS playback because currentAudioJob was null');
     }
-    const reader = new FileReader();
+    if (audioBlob instanceof Blob) {
-    reader.onload = function (e) {
+        const srcUrl = await getBase64Async(audioBlob);
-        const srcUrl = e.target.result;
         audioElement.src = srcUrl;
-    };
+    } else if (typeof audioBlob === 'string') {
-    reader.readAsDataURL(audioBlob);
+        audioElement.src = audioBlob;
+    } else {
+        throw `TTS received invalid audio data type ${typeof audioBlob}`;
+    }
     audioElement.addEventListener('ended', completeCurrentAudioJob);
     audioElement.addEventListener('canplay', () => {
         console.debug('Starting TTS playback');
@@ -419,11 +421,15 @@ function completeCurrentAudioJob() {
  * @param {Response} response
  */
 async function addAudioJob(response) {
-    const audioData = await response.blob();
+    if (typeof response === 'string') {
-    if (!audioData.type.startsWith('audio/')) {
+        audioJobQueue.push(response);
-        throw `TTS received HTTP response with invalid data format. Expecting audio/*, got ${audioData.type}`;
+    } else {
+        const audioData = await response.blob();
+        if (!audioData.type.startsWith('audio/')) {
+            throw `TTS received HTTP response with invalid data format. Expecting audio/*, got ${audioData.type}`;
+        }
+        audioJobQueue.push(audioData);
     }
-    audioJobQueue.push(audioData);
     console.debug('Pushed audio job to queue.');
 }
 
@@ -434,7 +440,7 @@ async function processAudioJobQueue() {
     }
     try {
         audioQueueProcessorReady = false;
-        currentAudioJob = audioJobQueue.pop();
+        currentAudioJob = audioJobQueue.shift();
         playAudioData(currentAudioJob);
         talkingAnimation(true);
     } catch (error) {
@@ -465,13 +471,25 @@ function saveLastValues() {
 }
 
 async function tts(text, voiceId, char) {
+    async function processResponse(response) {
+        // RVC injection
+        if (extension_settings.rvc.enabled && typeof window['rvcVoiceConversion'] === 'function')
+            response = await window['rvcVoiceConversion'](response, char, text);
+
+        await addAudioJob(response);
+    }
+
     let response = await ttsProvider.generateTts(text, voiceId);
 
-    // RVC injection
+    // If async generator, process every chunk as it comes in
-    if (extension_settings.rvc.enabled && typeof window['rvcVoiceConversion'] === 'function')
+    if (typeof response[Symbol.asyncIterator] === 'function') {
-        response = await window['rvcVoiceConversion'](response, char, text);
+        for await (const chunk of response) {
+            await processResponse(chunk);
+        }
+    } else {
+        await processResponse(response);
+    }
 
-    addAudioJob(response);
     completeTtsJob();
 }
 
@@ -658,6 +676,7 @@ function onSkipCodeblocksClick() {
 function onPassAsterisksClick() {
     extension_settings.tts.pass_asterisks = !!$('#tts_pass_asterisks').prop('checked');
     saveSettingsDebounced();
+    console.log("setting pass asterisks", extension_settings.tts.pass_asterisks)
 }
 
 //##############//
@@ -743,7 +762,7 @@ function getCharacters(unrestricted) {
     if (unrestricted) {
         const names = context.characters.map(char => char.name);
         names.unshift(DEFAULT_VOICE_MARKER);
-        return names;
+        return names.filter(onlyUnique);
     }
 
     let characters = [];
@@ -758,14 +777,13 @@ function getCharacters(unrestricted) {
         characters.push(context.name1);
         const group = context.groups.find(group => context.groupId == group.id);
         for (let member of group.members) {
-            // Remove suffix
+            const character = context.characters.find(char => char.avatar == member);
-            if (member.endsWith('.png')) {
+            if (character) {
-                member = member.slice(0, -4);
+                characters.push(character.name);
             }
-            characters.push(member);
         }
     }
-    return characters;
+    return characters.filter(onlyUnique);
 }
 
 function sanitizeId(input) {

system/st_files/style.css

@@ -0,0 +1,159 @@
+#tts_media_control {
+    /* order: 100; */
+    /*     width: 40px;
+    height: 40px;
+    margin: 0;
+    padding: 1px; */
+    outline: none;
+    border: none;
+    cursor: pointer;
+    /*     transition: 0.3s;
+    opacity: 0.7; */
+    display: flex;
+    align-items: center;
+    /* justify-content: center; */
+
+}
+
+#ttsExtensionMenuItem {
+    transition: 0.3s;
+    opacity: 0.7;
+}
+
+#ttsExtensionMenuItem:hover {
+    opacity: 1;
+    filter: brightness(1.2);
+}
+
+#tts_media_control:hover {
+    opacity: 1;
+    filter: brightness(1.2);
+}
+
+.voice_preview {
+    margin: 0.25rem 0.5rem;
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    gap: 0.5rem;
+}
+
+.voice_preview .voice_name {
+    text-align: left;
+    flex: 1;
+}
+
+.voice_preview .voice_lang {
+    width: 4rem;
+    text-align: left;
+}
+
+.voice_preview .fa-play {
+    cursor: pointer;
+}
+
+.tts-button {
+    margin: 0;
+    outline: none;
+    border: none;
+    cursor: pointer;
+    transition: 0.3s;
+    opacity: 0.7;
+    align-items: center;
+    justify-content: center;
+
+}
+
+.tts-button:hover {
+    opacity: 1;
+}
+
+.tts_block {
+    display: flex;
+    align-items: baseline;
+    column-gap: 5px;
+    flex-wrap: wrap;
+}
+
+.tts_custom_voices {
+    display: flex;
+    align-items: baseline;
+    gap: 5px;
+}
+
+.novel_tts_hints {
+    font-size: calc(0.9 * var(--mainFontSize));
+    display: flex;
+    flex-direction: column;
+    gap: 5px;
+    margin-bottom: 5px;
+}
+
+.at-settings-row {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    width: 100%;
+}
+
+.at-settings-option {
+    flex: 1;
+    margin: 0 10px;
+}
+
+.at-endpoint-option {
+    flex: 1; 
+    margin: 0 10px;
+    margin-right: 25px;
+    width: 38%;
+}
+
+.at-website-row {
+    display: flex;
+    justify-content: start;
+    align-items: center;
+    margin-top: 10px; 
+    margin-bottom: 10px;
+}
+
+.at-website-option {
+    flex: 1; 
+    margin-right: 10px; 
+    margin-left: 10px; 
+}
+
+.at-settings-separator {
+    margin-top: 10px;
+    margin-bottom: 10px;
+    padding: 18x;
+    font-weight: bold;
+    border-top: 1px solid #e1e1e1; /* Grey line */
+    border-bottom: 1px solid #e1e1e1; /* Grey line */
+    text-align: center;
+}
+
+.at-status-message {
+    flex: 1;
+    margin: 0 10px;
+}
+
+.at-model-endpoint-row {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    width: 100%;
+}
+
+.at-model-option, .endpoint-option {
+    flex: 1;
+    margin: 0 10px;
+    margin-left: 10px; 
+}
+
+.at-endpoint-option {
+    width: 38%;
+}
+
+#at-status_info {
+    color: lightgreen; 
+}

system/tts_diff/tts_diff.py

@@ -0,0 +1,269 @@
+import json
+import argparse
+from pathlib import Path
+import os
+import glob
+import platform
+import site
+import subprocess
+import sys
+# version 0.8
+
+def find_files_in_path_with_wildcard(pattern):
+    # Get the site-packages directory of the current Python environment
+    site_packages_path = site.getsitepackages()
+    found_paths = []
+    # Adjust the sub-directory based on the operating system
+    sub_directory = "nvidia/cublas"
+    if platform.system() == "Linux":
+        sub_directory = os.path.join(sub_directory, "lib")
+    else:
+        sub_directory = os.path.join(sub_directory, "bin")
+    
+    # Iterate over each site-packages directory (there can be more than one)
+    for directory in site_packages_path:
+        # Construct the search directory path
+        search_directory = os.path.join(directory, sub_directory)
+        # Use glob to find all files matching the pattern in this directory
+        for file_path in glob.glob(os.path.join(search_directory, pattern)):
+            if os.path.isfile(file_path):  # Ensure it's a file
+                found_paths.append(file_path)
+    return found_paths
+
+def detect_cublas():
+    # Use different file patterns based on the operating system
+    file_name_pattern = 'cublas64_11.*' if platform.system() == "Windows" else 'libcublas.so.11*'
+    found_paths = find_files_in_path_with_wildcard(file_name_pattern)
+    if found_paths:
+        print("[AllTalk TTSDiff] \033[94mCublas:\033[0m \033[92mDetected\033[0m", found_paths)
+    else:
+        print("[AllTalk TTSDiff] \033[94mCublas:\033[0m \033[91mERROR Not Detected\033[0m")
+        print("[AllTalk TTSDiff] \033[94mPlease install Cublas from the Nvidia CUDA Toolkit \033[0mhttps://developer.nvidia.com/cuda-downloads")
+        exit(1)
+        
+
+parser = argparse.ArgumentParser(description="Compare TTS output with the original text using detailed comparison.")
+parser.add_argument("--threshold", type=int, default=98, help="Similarity threshold for considering a match (default: 98)")
+parser.add_argument("--ttslistpath", help="Path to the ttsList.json file")
+parser.add_argument("--wavfilespath", help="Path to the wav outputs folder")
+args = parser.parse_args()
+
+if not args.ttslistpath:
+    parser.error("[AllTalk TTSDiff] Please specify the path to the ttslist.json file using the --ttslistpath argument.")
+if not args.wavfilespath:
+    parser.error("[AllTalk TTSDiff] Please specify the path to the wavs output folder file using the --wavfilespath argument.")
+
+json_file_path = Path(args.ttslistpath).resolve()
+wav_file_path = Path(args.wavfilespath).resolve()
+accuracy_threshold = args.threshold
+
+def disclaimer_text():
+    print(f"\nDESCRIPTION: ")
+    print(f"  This script is designed to assist in identifying bad TTS generation ID's from the TTS Generator.\n")
+    print(f"  This will download the Whisper Large-v2 model (if not already downloaded), and use it to compare your text to")
+    print(f"  to the generated TTS. If it finds an issue, it will flag the ID number that has the issue. This does NOT 100%")
+    print(f"  ensure that all TTS generated is perfect, sounds good etc, but it should flag up glaring issues and help with")
+    print(f"  the process of identifying bad TTS generations.")
+    print(f"\nNOTES ON USE: ")
+    print(f"  - Ensure AllTalk is running in its own command prompt/terminal.")
+    print(f"  - This script should be running in its own command prompt/terminal window in the same Python Environment that")
+    print(f"    AllTalk is running in.")
+    print(f"  - After generating your text in the TTS Generator, you need to `Export List to JSON` and save the file in the")
+    print(f"    same folder/directory this script is running from.")
+    print(f"  - You can use `--threshold` with a numerical value from 1 to 100 to set accuracy detection. For example it may")
+    print(f"    detect/transcribe `their` as `there`, which of course, both words sound the same. If you want 100% accuracy")
+    print(f"    you will set 100 as the threshold. You will more than likely want a number in the high 90's. Default is set")
+    print(f"    at 98, which gives a little flexibility.")
+    print(f"  - The JSON file must be named \033[93mttsList.json\033[0m")
+    print(f"  - When you have your ID list, go back into the TTS Generator, correct any lines and regenerate them. If you")
+    print(f"    want to re-test everthing again after re-generating, you will need to export your list again and re-run this")
+    print(f"    script again, against your newly exported JSON list.")
+    print(f"  - This requires access to \033[93mcublas64_11\033[0m, the same as Finetuning.")
+    print(f"    https://github.com/erew123/alltalk_tts/tree/main?#-important-requirements-cuda-118\n")
+    return
+
+try:
+    from faster_whisper import WhisperModel
+    import re
+    import string
+    import spacy
+    import torch
+    from fuzzywuzzy import fuzz
+except ImportError as e:
+    print(f"[AllTalk TTSDiff] ERROR STARTING SCRIPT:")
+    print("[AllTalk TTSDiff] An error occurred importing one or more required libraries.")
+    print("[AllTalk TTSDiff] Please ensure you have \033[94mactivated your Python environment \033[0mand \033[94minstalled all requirements.\033[0m")
+    print("[AllTalk TTSDiff] \033[94mMissing library:\033[93m", str(e), "\033[0m")
+    exit(1)
+
+# Attempt to load the spaCy model
+try:
+    # Load a medium-sized spaCy model, adjust to "en_core_web_md" or "en_core_web_lg" as needed
+    nlp = spacy.load("en_core_web_md")
+except Exception as e:
+    print(f"[AllTalk TTSDiff] ERROR STARTING SCRIPT:")
+    print("[AllTalk TTSDiff] Failed to load the spaCy language model.")
+    
+    # Attempt to download the spaCy model automatically
+    try:
+        print("[AllTalk TTSDiff] Attempting to automatically download the spaCy language model...")
+        subprocess.run([sys.executable, "-m", "spacy", "download", "en_core_web_md"], check=True)
+        
+        # Try to load the model again after downloading
+        nlp = spacy.load("en_core_web_md")
+        print("[AllTalk TTSDiff] Model downloaded and loaded successfully.")
+    except Exception as download_exception:
+        print("[AllTalk TTSDiff] Automatic download failed.")
+        print("[AllTalk TTSDiff] Please manually install the spaCy language model by running:")
+        print("[AllTalk TTSDiff] \033[93mpython -m spacy download en_core_web_md\033[0m")
+        exit(1)
+
+def texts_are_similar(text1, text2, threshold=0.8):
+    # Preliminary check for short texts
+    if len(text1) < 10 or len(text2) < 10:
+        ratio = fuzz.ratio(text1, text2) / 100.0  # Use FuzzyWuzzy for short texts
+        return ratio > threshold
+
+    # Process the texts through the NLP model for longer texts
+    doc1 = nlp(text1)
+    doc2 = nlp(text2)
+    
+    # Compute semantic similarity
+    similarity = doc1.similarity(doc2)
+    
+    return similarity >= threshold
+
+def normalize_text(text):
+    # Normalize or remove CRLF and other non-standard whitespaces
+    text = text.replace("\r\n", " ").replace("\n", " ").replace("\r", " ")
+    
+    # Convert to lowercase
+    text = text.lower()
+    
+    # Standardize and then remove quotation marks
+    text = text.replace("“", '"').replace("”", '"').replace("‘", "'").replace("’", "'")
+    text = text.translate(str.maketrans('', '', '"\''))
+    
+    # Remove all other punctuation except hyphens to preserve compound words
+    text = text.translate(str.maketrans('', '', string.punctuation.replace("-", "")))
+    
+    # Collapse any sequence of whitespace (including spaces) into a single space
+    text = re.sub(r'\s+', ' ', text)
+    
+    return text.strip()
+
+spoken_punctuation_mapping = {
+    "dot": ".",
+    "comma": ",",
+    # Add more mappings as needed
+}
+
+def contains_spoken_punctuation(transcribed_text):
+    for spoken, punctuation in spoken_punctuation_mapping.items():
+        if spoken in transcribed_text:
+            return True
+    return False
+
+def detailed_comparison(original_text, transcribed_text, threshold=98):
+    original_clean = normalize_text(original_text)
+    transcribed_clean = normalize_text(transcribed_text)
+    ratio = fuzz.ratio(original_clean, transcribed_clean)
+    return ratio > threshold
+
+def access_local_file(url):
+    # Extract the file name from the URL
+    file_name = url.split('/')[-1]
+    # Construct the full path to the file on disk
+    file_path = wav_file_path / file_name
+    # Check if the file exists
+    if not file_path.exists():
+        print(f"[AllTalk TTSDiff] File does not exist: {file_path}")
+        return None
+    return file_path
+
+def transcribe_and_compare(file_url, original_text, model, item_id, flagged_ids):
+    audio_file_path = access_local_file(file_url)
+    if audio_file_path is None:
+        print(f"[AllTalk TTSDiff] Could not access local file for URL: {file_url}")
+        return  # Skip this file if it can't be accessed
+
+    try:
+        segments, info = model.transcribe(str(audio_file_path), beam_size=5)
+        transcribed_text = " ".join([segment.text for segment in segments])
+    except Exception as e:
+        print(f"[AllTalk TTSDiff] Error transcribing file {audio_file_path}: {e}")
+        return  # Skip this file if transcription fails
+    
+    segments, info = model.transcribe(str(audio_file_path), beam_size=5)
+    transcribed_text = " ".join([segment.text for segment in segments])
+    
+    # Normalize texts for comparison
+    original_text_normalized = normalize_text(original_text)
+    transcribed_text_normalized = normalize_text(transcribed_text)
+    
+    # Enhanced comparison using detailed fuzzy matching
+    is_detailed_match = detailed_comparison(original_text_normalized, transcribed_text_normalized, args.threshold)
+    has_spoken_punctuation = contains_spoken_punctuation(transcribed_text_normalized) and not contains_spoken_punctuation(original_text_normalized)
+    
+    # Adjust is_match based on detailed comparison and spoken punctuation check
+    is_match = is_detailed_match and not has_spoken_punctuation
+    
+    # Only log and flag IDs for review if there's a mismatch or detected issues
+    if not is_match:
+        #print(f"\033[93mMismatch Detected - ID:\033[0m {item_id}")
+        print(f"[AllTalk TTSDiff] \033[93mText:\033[0m {original_text}")
+        print(f"[AllTalk TTSDiff] \033[91mTTS :\033[0m{transcribed_text}")
+        if has_spoken_punctuation:
+            print(f"[AllTalk TTSDiff] Note: Potential incorrect spoken punctuation detected.")
+        flagged_ids.append(item_id)  # Track the ID for review
+
+def main():
+    print(f"[AllTalk TTSDiff]")
+    print(f"[AllTalk TTSDiff]")
+    print("[AllTalk TTSDiff] \033[92mStarting Compare of Text vs TTS:\033[0m")
+    print("[AllTalk TTSDiff]")
+    print("[AllTalk TTSDiff] \033[94mJSON file  :\033[92m", json_file_path, "\033[0m")
+    print("[AllTalk TTSDiff] \033[94mWAV files  :\033[92m", wav_file_path, "\033[0m")
+    print("[AllTalk TTSDiff] \033[94mAccuracy   :\033[92m", accuracy_threshold, "%", "\033[0m")
+    detect_cublas()
+    model_size = "large-v2"
+    device = "cuda" if torch.cuda.is_available() else "cpu"
+    model = WhisperModel(model_size, device=device, compute_type="float32")
+    
+    flagged_ids = []  # Initialize the list to track IDs needing review
+
+    try:
+        with open(json_file_path, "r", encoding="utf-8") as f:
+            tts_list = json.load(f)
+    except FileNotFoundError:
+        print(f"[AllTalk TTSDiff] ERROR STARTING SCRIPT:")
+        print("[AllTalk TTSDiff] Error: \033[93mttsList.json\033[0m file not found.")
+        print("[AllTalk TTSDiff] Please ensure you follow these steps:")
+        print("[AllTalk TTSDiff] After generating your text in the TTS Generator, you need to `Export List to JSON` and save the file in the")
+        print("[AllTalk TTSDiff] same folder/directory this script is running from.")
+        print("[AllTalk TTSDiff] The JSON file must be named \033[93mttsList.json\033[0m")
+        exit(1)
+
+    print(f"[AllTalk TTSDiff]")
+
+    for item in tts_list:
+        print(f"[AllTalk TTSDiff] Processing ID: {item['id']}")
+        transcribe_and_compare(item['fileUrl'], item['text'], model, item['id'], flagged_ids)
+    
+    print(f"[AllTalk TTSDiff]")
+    # Print summary information at the end
+    if flagged_ids:
+        print(f"[AllTalk TTSDiff] \033[94mIDs needing review:\033[0m", ', '.join(map(str, flagged_ids)))
+        print(f"[AllTalk TTSDiff]")
+        print(f"[AllTalk TTSDiff] Review ID's and correct any lines by editing & regenerating as necessary.")
+    else:
+        print("[AllTalk TTSDiff] \033[94mIDs needing review:\033[0m No issues detected.")
+    summary_data = {
+        "flagged_ids": flagged_ids
+    }
+    with open(wav_file_path / "analysis_summary.json", "w") as summary_file:
+        json.dump(summary_data, summary_file)
+    print("[AllTalk TTSDiff]")
+
+if __name__ == "__main__":
+    main()

system/tts_generator/tts_generator.html

@@ -6,7 +6,6 @@
     <meta name="viewport" content="width=device-width, initial-scale=1.0">
     <link rel="icon" href="./favicon.ico" type="image/x-icon">
     <title>AllTalk TTS Generator</title>
-
     <style>
         body,
         textarea,
@@ -367,6 +366,11 @@
             /* Other styles for highlighting */
         }
 
+        .problematic {
+            background-color: rgb(179, 104, 104);
+            /* Or any light red color of your choice */
+        }
+
         .highlight-count {
             color: #ff7300;
             /* Change to the color you want */
@@ -391,6 +395,13 @@
             font-size: 0.8em;
         }
 
+        /*Disable and enable buttons*/
+        button:disabled {
+            background-color: #919191;
+            color: #666;
+            cursor: not-allowed;
+        }
+
         .edit-input {
             /* Styles for the edit input field */
             margin-top: 5px;
@@ -563,7 +574,7 @@
         <div class="setting slider-setting-wrapper">
             <label for="volume">Volume:</label>
             <div class="slider-container">
-                <input type="range" id="volume" name="volume" class="slider-setting" min="0.0" max="1.0" step="0.1"
+                <input type="range" id="volume" name="volume" class="slider-setting" min="0.1" max="1.0" step="0.1"
                     value="0.8" oninput="updateVolumeValue(this.value)">
                 <span id="volumeValue">0.8</span>
             </div>
@@ -596,14 +607,17 @@
 <div id="ttsListSection" class="tts-list-section">
     <h2>Generated TTS List</h2>
     <input type="number" id="startFromId" value="1" min="1" autocomplete="off" style="width: 60px; margin-right: 10px;">
-    <button id="playPauseList" onclick="togglePlayPauseList()" style="margin-right: 10px;">Play List</button>
+    <button class="conditional-btn" id="playPauseList" onclick="togglePlayPauseList()" style="margin-right: 10px;">Play
-    <button onclick="stopPlayback()" style="margin-right: 10px;">Stop Playback</button>
+        List</button>
-    <button id="clearList" onclick="clearTTSList()" style="margin-right: 10px;">Clear List</button>
+    <button class="conditional-btn" onclick="stopPlayback()" style="margin-right: 10px;">Stop Playback</button>
-    <button onclick="exportToJSON()" style="margin-right: 10px;">Export List as JSON</button>
+    <button class="conditional-btn" id="clearList" onclick="clearTTSList()" style="margin-right: 40px;">Clear
-    <button onclick="importFromJSON()" style="margin-right: 10px;">Import JSON to list</button>
+        List</button>
-    <button onclick="exportMP3()" style="margin-right: 10px;">Export to Wav</button>
+    <button onclick="importFromJSON()" style="margin-right: 10px;">Import JSON</button>
-    <label for="splitEvery">Split every:</label>
+    <button class="conditional-btn" onclick="exportToJSON()" style="margin-right: 10px;">Export JSON</button>
-    <select id="splitEvery" name="splitEvery">
+    <button class="conditional-btn" id="exportsrttitles" style="margin-right: 10px;">Export SRT</button>
+    <button class="conditional-btn" onclick="exportMP3()" style="margin-right: 10px;">Export to WAV</button>
+    <label for="splitEvery">Export Split:</label>
+    <select id="splitEvery" name="splitEvery" style="margin-right: 40px;">
         <option value="100">100</option>
         <option value="200">200</option>
         <option value="300">300</option>
@@ -611,22 +625,38 @@
         <option value="500" selected>500</option>
         <option value="600">600</option>
     </select>
+    <button class="conditional-btn" id="saveAndAnalyzeTTS" style="margin-right: 10px;">Analyze TTS</button>
+    <label for="AnalyzeAccuracy">% Accuracy:</label>
+    <select id="AnalyzeAccuracy" name="AnalyzeAccuracy" style="margin-right: 10px;">
+        <option value="99">99</option>
+        <option value="98" selected>98</option>
+        <option value="97">97</option>
+        <option value="96">96</option>
+        <option value="95">95</option>
+        <option value="94">94</option>
+        <option value="93">93</option>
+        <option value="92">92</option>
+        <option value="91">91</option>
+        <option value="90">90</option>
+    </select>
     <div id="paginationControls">
-        <button id="prevPage" class="pagination-btn">Previous</button>
+        <button class="conditional-btn pagination-btn" id="prevPage">Previous</button>
         <span id="pageIndicator">Page 1 of X</span>
-        <button id="nextPage" class="pagination-btn">Next</button>
+        <button class="conditional-btn pagination-btn" id="nextPage">Next</button>
-        <input type="number" id="jumpToPage" class="pagination-input" min="1" placeholder="Jump to Page">
+        <input type="number" id="jumpToPage" class="conditional-btn pagination-input" min="1"
-        <button id="jumpButton" class="pagination-btn">Go</button>
+            placeholder="Jump to Page">
+        <button class="conditional-btn pagination-btn" id="jumpButton">Go</button>
     </div>
     <div id="ttsListDisplay" class="tts-list-display">
         <!-- Dynamically populated TTS details will go here -->
     </div>
     <div id="paginationControls1">
-        <button id="prevPage1" class="pagination-btn">Previous</button>
+        <button class="conditional-btn pagination-btn" id="prevPage1">Previous</button>
         <span id="pageIndicator1">Page 1 of X</span>
-        <button id="nextPage1" class="pagination-btn">Next</button>
+        <button class="conditional-btn pagination-btn" id="nextPage1">Next</button>
-        <input type="number" id="jumpToPage1" class="pagination-input" min="1" placeholder="Jump to Page">
+        <input type="number" id="jumpToPage1" class="conditional-btn pagination-input" min="1"
-        <button id="jumpButton1" class="pagination-btn">Go</button>
+            placeholder="Jump to Page">
+        <button class="conditional-btn pagination-btn" id="jumpButton1">Go</button>
     </div>
 </div>
 </div>
@@ -637,58 +667,56 @@
     import Crunker from './crunker.esm.js';
 
     async function exportMP3() {
-    const crunker = new Crunker();
+        const crunker = new Crunker();
-    const fileUrls = ttsList.map(item => item.fileUrl);
+        const fileUrls = ttsList.map(item => item.fileUrl);
-    const splitEvery = parseInt(document.getElementById('splitEvery').value) || 500;
+        const splitEvery = parseInt(document.getElementById('splitEvery').value) || 500;
-    const progressIndicator = document.getElementById('progressIndicator');
+        const progressIndicator = document.getElementById('progressIndicator');
-    const progressText = document.getElementById('progressText');
+        const progressText = document.getElementById('progressText');
 
-    if (progressIndicator && progressText) {
-        progressIndicator.style.display = 'block';
-        progressText.textContent = 'Starting export...';
-    }
-
-    try {
-        for (let i = 0; i < fileUrls.length; i += splitEvery) {
-            const currentBatch = Math.floor(i / splitEvery) + 1;
-            const totalBatches = Math.ceil(fileUrls.length / splitEvery);
-            if (progressText) {
-                progressText.textContent = `Processing batch ${currentBatch} of ${totalBatches}`;
-                progressText.style.color = 'black'; // Set the text color to black
-            }
-
-            const batchUrls = fileUrls.slice(i, Math.min(i + splitEvery, fileUrls.length));
-            const audioBuffers = await crunker.fetchAudio(...batchUrls);
-            const combined = crunker.concatAudio(audioBuffers);
-            const output = crunker.export(combined, "audio/wav");
-            crunker.download(output.blob, `combined_audio_part_${currentBatch}.wav`);
-
-            // Clear audioBuffers array to free up memory
-            audioBuffers.length = 0;
-
-            console.log(`Exported batch ${currentBatch} of ${totalBatches}`);
-        }
-    } catch (error) {
-        console.error('Error combining files:', error);
-        if (progressText) {
-            progressText.textContent = 'Error during export!';
-        }
-    } finally {
         if (progressIndicator && progressText) {
-            progressIndicator.style.display = 'none';
+            progressIndicator.style.display = 'block';
-            progressText.textContent = ''; // Reset progress text
+            progressText.textContent = 'Starting export...';
+        }
+
+        try {
+            for (let i = 0; i < fileUrls.length; i += splitEvery) {
+                const currentBatch = Math.floor(i / splitEvery) + 1;
+                const totalBatches = Math.ceil(fileUrls.length / splitEvery);
+                if (progressText) {
+                    progressText.textContent = `Processing batch ${currentBatch} of ${totalBatches}`;
+                    progressText.style.color = 'black'; // Set the text color to black
+                }
+
+                const batchUrls = fileUrls.slice(i, Math.min(i + splitEvery, fileUrls.length));
+                const audioBuffers = await crunker.fetchAudio(...batchUrls);
+                const combined = crunker.concatAudio(audioBuffers);
+                const output = crunker.export(combined, "audio/wav");
+                crunker.download(output.blob, `combined_audio_part_${currentBatch}.wav`);
+
+                // Clear audioBuffers array to free up memory
+                audioBuffers.length = 0;
+
+                console.log(`Exported batch ${currentBatch} of ${totalBatches}`);
+            }
+        } catch (error) {
+            console.error('Error combining files:', error);
+            if (progressText) {
+                progressText.textContent = 'Error during export!';
+            }
+        } finally {
+            if (progressIndicator && progressText) {
+                progressIndicator.style.display = 'none';
+                progressText.textContent = ''; // Reset progress text
+            }
         }
     }
-}
 
-window.exportMP3 = exportMP3;
+    window.exportMP3 = exportMP3;
 
 
 
 </script>
 
-
-
 <script>
 
     /////////////////////////////////////////
@@ -705,12 +733,16 @@ window.exportMP3 = exportMP3;
     let isStreamInitialized = false;
     let characterVoice = document.getElementById("character_voice_gen").value;
     let ttsList = []; // Global list to store TTS details
+    let problematicIds = []; // Global list to store ProblematicIDs
     const howlInstances = {};
     let listPlaying = false;
     let listPlayIndex = 0;
     let currentPage = 1;
     let itemsPerPage = 100; // Adjust as needed
-
+    var protocol = window.location.protocol;
+    var host = window.location.hostname;
+    var port = window.location.port;
+    var baseUrl = protocol + "//" + host + (port ? ':' + port : '');
 
     ////////////////////////////////////
     //////// Progress Indicator ////////
@@ -757,6 +789,123 @@ window.exportMP3 = exportMP3;
     }
 
 
+    //////////////////////////////////
+    //////// Compare TTS list ////////
+    //////////////////////////////////
+
+    document.getElementById('saveAndAnalyzeTTS').addEventListener('click', async function () {
+        showProgress(); // Call this function to show the progress indicator at the start of the process
+        try {
+            // Serialize the ttsList data structure to a JSON string
+            const ttsDataJSON = JSON.stringify(ttsList);
+
+            // Send the serialized ttsList to the '/save-tts-data' endpoint
+            const saveResponse = await fetch('/api/save-tts-data', {
+                method: 'POST',
+                headers: { 'Content-Type': 'application/json' },
+                body: ttsDataJSON  // Use the serialized JSON string as the request body
+            });
+
+            const saveData = await saveResponse.json();
+
+            // Log the server's response
+            console.log(saveData.message);
+
+            // If the server confirms the data was saved successfully, trigger analysis
+            if (saveData.message === "Data saved successfully") {
+                const accuracy = parseInt(document.getElementById('AnalyzeAccuracy').value) || 98;
+                const analyzeResponse = await fetch(`/api/trigger-analysis?threshold=${accuracy}`);
+                const analyzeData = await analyzeResponse.json();
+                if (analyzeData.summary.flagged_ids && analyzeData.summary.flagged_ids.length > 0) {
+                    // Update the global problematicIds array with the new flagged IDs
+                    problematicIds = analyzeData.summary.flagged_ids;
+                    console.log("Initial problematicIds from analysis:", problematicIds); // Debugging: Check the initial array after receiving analysis results
+
+                    // Call highlightProblems with the array of flagged IDs
+                    highlightProblems(problematicIds);
+                } else {
+                    console.log("No problematic IDs or error retrieving data.");
+
+                    // Clear the problematicIds array since there are no problematic IDs
+                    problematicIds = [];
+                    highlightProblems(problematicIds);
+                }
+                // Optionally, you might want to provide feedback in the UI about the analysis status
+            } else {
+                throw new Error('Data save was unsuccessful.');
+            }
+        } catch (error) {
+            // Handle any errors that occurred during the fetch process
+            console.error('Error:', error);
+        } finally {
+            hideProgress(); // Hide the progress indicator when the process is complete or if an error occurs
+        }
+    });
+
+    function highlightProblems(idsToHighlight) {
+        // First, clear all previous 'problematic' highlights
+        document.querySelectorAll('.problematic').forEach(item => {
+            item.classList.remove('problematic');
+        });
+
+        // Now, highlight all the IDs that should still be problematic
+        idsToHighlight.forEach(id => {
+            const itemElement = document.getElementById(`item-${id}`);
+            if (itemElement) {
+                itemElement.classList.add('problematic');
+            }
+        });
+    }
+
+    /////////////////////////////////
+    //////// Export SRT Subs ////////
+    /////////////////////////////////
+
+    document.getElementById('exportsrttitles').addEventListener('click', async function () {
+        showProgress(); // Call this function to show the progress indicator at the start of the process
+        try {
+            const ttsDataJSON = JSON.stringify(ttsList);
+
+            const saveResponse = await fetch('/api/save-tts-data', {
+                method: 'POST',
+                headers: { 'Content-Type': 'application/json' },
+                body: ttsDataJSON
+            });
+            const saveData = await saveResponse.json();
+            console.log(saveData.message);
+
+            if (saveData.message === "Data saved successfully") {
+                // Trigger the SRT generation and handle file download
+                const analyzeResponse = await fetch(`/api/srt-generation`);
+                if (analyzeResponse.ok) {
+                    // Create a temporary anchor element
+                    const a = document.createElement('a');
+                    // Create a blob URL from the response
+                    const blob = await analyzeResponse.blob();
+                    const url = window.URL.createObjectURL(blob);
+                    a.href = url;
+                    // Set the download name for the file
+                    a.download = "subtitles.srt";
+                    // Append the anchor to the document
+                    document.body.appendChild(a);
+                    // Trigger the download
+                    a.click();
+                    // Clean up by removing the anchor element and revoking the blob URL
+                    window.URL.revokeObjectURL(url);
+                    a.remove();
+                } else {
+                    console.log("Error running SRT generation.");
+                }
+            } else {
+                throw new Error('Data save was unsuccessful.');
+            }
+        } catch (error) {
+            console.error('Error:', error);
+        } finally {
+            hideProgress(); // Hide the progress indicator when the process is complete or if an error occurs
+        }
+    });
+
     ///////////////////////////////////////////
     //////// TTS Generation Activities ////////
     ///////////////////////////////////////////
@@ -775,7 +924,14 @@ window.exportMP3 = exportMP3;
             let cleanedText = textInput.replace(/ \- | \– /g, ' ');
             cleanedText = cleanedText.replace(/%/g, " percent");
             cleanedText = cleanedText.replace(/\.\s'/g, ".'");
+            cleanedText = cleanedText.replace(/(?<!\w)[']+(?!\w)/g, '');
+            cleanedText = cleanedText.replace(/"/g, "'");
             cleanedText = cleanedText.replace(/[^a-zA-Z0-9\s\.,;:!?\-\'"À-ÿ\u0400-\u04FF\u0150\u0151\u0170\u0171]\$/g, '');
+            // Replace dash followed by space with comma and space
+            cleanedText = cleanedText.replace(/-\s/g, '; ');
+            // Replace double dash "--" with semicolon "; "
+            cleanedText = cleanedText.replace(/--/g, '; ');
+
             let sentences = cleanedText.split(/(?<=[.!?:])\s+/);
 
             if (!sentences) {
@@ -851,7 +1007,7 @@ window.exportMP3 = exportMP3;
                 autoplay_volume: volume
             });
 
-            const response = await fetch("http://127.0.0.1:7851/api/tts-generate", {
+            const response = await fetch(baseUrl + "/api/tts-generate", {
                 method: "POST",
                 headers: {
                     "Content-Type": "application/x-www-form-urlencoded",
@@ -884,6 +1040,7 @@ window.exportMP3 = exportMP3;
 
                 // Refresh the list display to reflect the new pagination state
                 updateUIList();
+                updateButtonStyles()
 
                 // Update the audio queue and UI
                 const sound = new Howl({
@@ -900,11 +1057,50 @@ window.exportMP3 = exportMP3;
                     playNextLocalChunk();
                 }
             }
+            return result
         } catch (error) {
             console.error("Error in generateAndCache:", error);
         }
     }
 
+    async function regenerateTTS(id) {
+        console.log("Before regeneration:", problematicIds); // Debugging: Check the array before regeneration
+
+        // Find the details of the TTS to regenerate
+        const ttsDetails = ttsList.find(item => item.id === id);
+        if (!ttsDetails) return;
+
+        // Retrieve the current volume setting from the HTML input
+        const currentVolume = parseFloat(document.getElementById("volume").value);
+
+        // Call the TTS generation with the same details and current volume
+        const result = await generateAndCache(ttsDetails.text, ttsDetails.language, ttsDetails.characterVoice, currentVolume, id);
+
+        if (result && result.status === "generate-success") {
+            const index = problematicIds.indexOf(id);
+            if (index !== -1) {
+                problematicIds.splice(index, 1); // Remove the id from problematicIds
+                console.log(`ID ${id} removed from problematicIds. Updated array:`, problematicIds); // Debugging: Check the array after removing an ID
+            }
+
+            // Find the element for this ID and remove the 'problematic' class
+            const problemItemElement = document.getElementById(`item-${id}`);
+            if (problemItemElement) {
+                problemItemElement.classList.remove('problematic');
+            }
+        } else {
+            console.log(`Regeneration failed for ID ${id}.`, result); // If regeneration fails, log the result
+        }
+
+        // Refresh the UI list to reflect the new data
+        updateUIList(); // Assume this function exists and it updates the display list
+        // Reapply highlighting to the items still in problematicIds
+        highlightProblems(problematicIds);
+        console.log("After regeneration and re-highlighting:", problematicIds); // Debugging: Check the array after re-highlighting
+    }
+
+
+
     async function generateWithoutPlayback(chunk, language, characterVoice, id = null) {
         // Perform the generation process as usual but do not play the audio
         try {
@@ -921,7 +1117,7 @@ window.exportMP3 = exportMP3;
                 autoplay: "false",
                 autoplay_volume: "0.1"
             });
-            const response = await fetch("http://127.0.0.1:7851/api/tts-generate", {
+            const response = await fetch(baseUrl + "/api/tts-generate", {
                 method: "POST",
                 headers: {
                     "Content-Type": "application/x-www-form-urlencoded",
@@ -970,19 +1166,6 @@ window.exportMP3 = exportMP3;
         }
     }
 
-
-    async function regenerateTTS(id) {
-        // Find the details of the TTS to regenerate
-        const ttsDetails = ttsList.find(item => item.id === id);
-        if (!ttsDetails) return;
-
-        // Retrieve the current volume setting from the HTML input
-        const currentVolume = parseFloat(document.getElementById("volume").value);
-
-        // Call the TTS generation with the same details and current volume
-        await generateAndCache(ttsDetails.text, ttsDetails.language, ttsDetails.characterVoice, currentVolume, id);
-    }
-
     function updateUIList() {
         const listDisplay = document.getElementById('ttsListDisplay');
         if (!listDisplay) {
@@ -1027,7 +1210,9 @@ window.exportMP3 = exportMP3;
         document.getElementById('pageIndicator').innerText = `Page ${currentPage} of ${lastPage}`;
         document.getElementById('pageIndicator1').innerText = `Page ${currentPage} of ${lastPage}`;
         document.getElementById('prevPage').disabled = (currentPage === 1);
+        document.getElementById('prevPage1').disabled = (currentPage === 1);
         document.getElementById('nextPage').disabled = (currentPage === lastPage);
+        document.getElementById('nextPage1').disabled = (currentPage === lastPage);
     }
 
 
@@ -1051,7 +1236,7 @@ window.exportMP3 = exportMP3;
                 autoplay_volume: volume,
             });
 
-            const response = await fetch("http://127.0.0.1:7851/api/tts-generate", {
+            const response = await fetch(baseUrl + "/api/tts-generate", {
                 method: "POST",
                 headers: {
                     "Content-Type": "application/x-www-form-urlencoded",
@@ -1081,8 +1266,9 @@ window.exportMP3 = exportMP3;
             });
 
             // Reinitialize currentSound every time
+            var srcUrl = baseUrl + "/tts-demo-request?" + queryParams.toString();
             currentSound = new Howl({
-                src: [`http://127.0.0.1:7851/tts-demo-request?${queryParams.toString()}`],
+                src: [srcUrl],
                 volume: volume,
                 html5: true // Force HTML5 Audio to enable streaming
             });
@@ -1141,7 +1327,7 @@ window.exportMP3 = exportMP3;
             const formData = new FormData();
             formData.append("voice", characterVoice);
 
-            const response = await fetch("http://127.0.0.1:7851/api/previewvoice/", {
+            const response = await fetch(baseUrl + "/api/previewvoice/", {
                 method: "POST",
                 body: formData,
             });
@@ -1237,7 +1423,7 @@ window.exportMP3 = exportMP3;
         } else {
             // In chunk mode, toggle the chunk play/pause
             if (isPlaying && currentlyPlayingIndex > 0) {
-                let currentSound = audioQueue[currentlyPlayingIndex - 1];
+                let currentSound = audioQueue[currentlyPlayingIndex - 3];
                 if (currentSound.playing()) {
                     currentSound.pause();
                 } else {
@@ -1264,16 +1450,6 @@ window.exportMP3 = exportMP3;
         }
     }
 
-
-    function stopPlayback() {
-        Howler.stop();
-        audioQueue = [];
-        // Unload each sound in the queue
-        audioQueue.forEach(sound => sound.unload());
-        currentlyPlayingIndex = 0;
-        isPlaying = false;
-    }
-
     function updatePlaybackButton() {
         const playbackButton = document.querySelector(".toggle-playback");
 
@@ -1448,38 +1624,64 @@ window.exportMP3 = exportMP3;
         if (pageIndicator1) {
             pageIndicator1.textContent = 'Page 1 of 1';
         }
+        updateButtonStyles()
+    }
+
+
+    function encodeHtmlEntities(text) {
+        return text.replace(/"/g, '&quot;').replace(/'/g, '&#39;');
+    }
+
+    function decodeHtmlEntities(text) {
+        var textArea = document.createElement('textarea');
+        textArea.innerHTML = text;
+        return textArea.value;
     }
 
     function editText(id) {
+        console.log(`Initiating edit for ID: ${id}`);
+        console.log("Before edit, ttsList:", JSON.parse(JSON.stringify(ttsList)));
         // First, check if we're already editing and should save the current value
         const existingInput = document.getElementById(`edit-input-${id}`);
         if (existingInput) {
+            console.log(`Existing input found for ID: ${id}, triggering blur event to save.`);
             existingInput.blur(); // This will trigger the blur event, saving the current value
             return; // Don't proceed further since we're just saving the existing input
         }
-
         const entryDiv = document.getElementById(`item-${id}`);
         if (!entryDiv) {
             console.error(`Item with ID ${id} not found.`);
             return;
         }
-
         const textPreviewSpan = entryDiv.querySelector('.text-preview');
-        const fullText = textPreviewSpan.getAttribute('data-fulltext');
+        // Retrieve the full text and immediately encode it
+        let fullText = textPreviewSpan.getAttribute('data-fulltext');
+        fullText = decodeHtmlEntities(fullText);
+        //fullText = encodeHtmlEntities(fullText);
+        console.log(`Current full text for ID: ${id} is: ${fullText}`);
 
         // Create an input field for editing
         const inputField = document.createElement('input');
         inputField.type = 'text';
+        // The value is already encoded at this point
         inputField.value = fullText;
         inputField.classList.add('edit-input');
         inputField.id = `edit-input-${id}`; // Assign an ID to the input field for later access
 
         // Handle saving the edited text
         inputField.addEventListener('blur', function () {
+            console.log(`Saving edited text for ID: ${id}: ${this.value}`);
             // Update the text in the ttsList array and the data-fulltext attribute
             const itemIndex = ttsList.findIndex(item => item.id === id);
             if (itemIndex !== -1) {
+                console.log(`Updating ttsList with new text for ID: ${id}`);
                 ttsList[itemIndex].text = this.value;
+                // If this item is in the list of problematic IDs, un-highlight it
+                if (problematicIds.includes(id)) {
+                    console.log(`ID: ${id} is problematic, removing highlight class`);
+                    const problemItemElement = document.getElementById(`item-${id}`);
+                    problemItemElement.classList.remove('problematic');
+                }
             }
 
             // Create a new span element to replace the input field
@@ -1491,15 +1693,18 @@ window.exportMP3 = exportMP3;
             newTextPreviewSpan.textContent = displayText;
 
             // Replace the input field with the new span
+            console.log(`Replacing input field with text span for ID: ${id}`);
             inputField.replaceWith(newTextPreviewSpan);
         });
 
         // Replace the span with the input field
+        console.log(`Replacing text span with input field for ID: ${id}`);
         textPreviewSpan.replaceWith(inputField);
 
         // Focus the input field and select its content to facilitate editing
         inputField.focus();
         inputField.select();
+        console.log("After edit, ttsList:", JSON.parse(JSON.stringify(ttsList)));
     }
 
     ////////////////////////////////////////
@@ -1531,63 +1736,72 @@ window.exportMP3 = exportMP3;
     }
 
     function importFromJSON() {
-    const input = document.createElement('input');
+        const input = document.createElement('input');
-    input.type = 'file';
+        input.type = 'file';
-    input.accept = '.json';
+        input.accept = '.json';
 
-    input.onchange = async (e) => {
+        input.onchange = async (e) => {
-        const file = e.target.files[0];
+            const file = e.target.files[0];
-        if (file) {
+            if (file) {
-            const reader = new FileReader();
+                const reader = new FileReader();
-            reader.readAsText(file);
+                reader.readAsText(file);
-            reader.onload = (e) => {
+                reader.onload = (e) => {
-                try {
+                    try {
-                    const jsonData = JSON.parse(e.target.result);
+                        const jsonData = JSON.parse(e.target.result);
 
-                    ttsList = jsonData;
+                        ttsList = jsonData;
-                    currentPage = 1; // Reset to the first page
+                        currentPage = 1; // Reset to the first page
-                    updateUIList();
+                        updateUIList();
-                    updatePaginationControls(); // Update pagination controls
+                        updateButtonStyles()
-                } catch (error) {
+                        updatePaginationControls(); // Update pagination controls
-                    console.error('Error parsing JSON:', error);
+                    } catch (error) {
-                }
+                        console.error('Error parsing JSON:', error);
-            };
+                    }
-        }
+                };
-    };
+            }
+        };
 
-    input.click();
+        input.click();
-}
-
-function updatePaginationControls() {
-    const lastPage = Math.ceil(ttsList.length / itemsPerPage);
-    const pageIndicator = document.getElementById('pageIndicator');
-    if (pageIndicator) {
-        pageIndicator.textContent = `Page ${currentPage} of ${lastPage}`;
     }
 
-    const prevPageButton = document.getElementById('prevPage');
+    function updatePaginationControls() {
-    if (prevPageButton) {
+        const lastPage = Math.ceil(ttsList.length / itemsPerPage);
-        prevPageButton.disabled = (currentPage === 1);
+        const prevPageButtons = document.querySelectorAll('.pagination-btn[id^="prevPage"]');
+        const nextPageButtons = document.querySelectorAll('.pagination-btn[id^="nextPage"]');
+
+        // Update page indicators and buttons
+        prevPageButtons.forEach(btn => btn.disabled = (currentPage === 1));
+        nextPageButtons.forEach(btn => btn.disabled = (currentPage === lastPage));
+
+        // Assuming the page indicators are consistent with button IDs
+        document.querySelectorAll('span[id^="pageIndicator"]').forEach(indicator => {
+            indicator.textContent = `Page ${currentPage} of ${lastPage}`;
+        });
+
+        // Update 'Jump to Page' inputs and 'Go' buttons
+        const jumpToPageInputs = document.querySelectorAll('.pagination-input[id^="jumpToPage"]');
+        const jumpButtons = document.querySelectorAll('.pagination-btn[id^="jumpButton"]');
+        const moreThanOnePage = lastPage > 1;
+
+        jumpToPageInputs.forEach(input => input.disabled = !moreThanOnePage);
+        jumpButtons.forEach(button => button.disabled = !moreThanOnePage);
     }
 
-    const nextPageButton = document.getElementById('nextPage');
-    if (nextPageButton) {
-        nextPageButton.disabled = (currentPage === lastPage);
-    }
-
-    // Similar updates for any additional pagination controls if you have them
-    // ...
-}
-
-
-
-
-
-
     //////////////////////////////
     //////// Other Things ////////
     //////////////////////////////
 
+    function updateButtonStyles() {
+        const ttsList = document.getElementById('ttsListDisplay');
+        const buttons = document.querySelectorAll('.conditional-btn');
+
+        if (!ttsList.hasChildNodes()) { // Check if the list is empty
+            buttons.forEach(button => button.disabled = true); // Disable buttons
+        } else {
+            buttons.forEach(button => button.disabled = false); // Enable buttons
+        }
+    }
+
     function updateLanguage(language) {
         // Update the language in your application logic
         console.log("Selected language:", language);
@@ -1633,7 +1847,7 @@ function updatePaginationControls() {
 
     async function fetchVoices() {
         try {
-            const response = await fetch("http://127.0.0.1:7851/api/voices");
+            const response = await fetch(baseUrl + "/api/voices");
             const result = await response.json();
             const characterVoiceSelect = document.getElementById("character_voice_gen");
 
@@ -1660,10 +1874,10 @@ function updatePaginationControls() {
     function stopPlayback() {
         Howler.stop();
         audioQueue = [];
-        currentlyPlayingIndex = 0;
+        //currentlyPlayingIndex = 0;
         isPlaying = false;
-        highlightPlaying(null);
+        //highlightPlaying(null);
-        document.getElementById('startFromId').value = '1'; // Reset the startFromId input to 1
+        //document.getElementById('startFromId').value = '1'; // Reset the startFromId input to 1
     }
 
     function displayOutput(result) {
@@ -1833,6 +2047,7 @@ function updatePaginationControls() {
     function setupPage() {
         // Fetch voices or other initial data
         fetchVoices();
+        updateButtonStyles()
 
         // Check for dark mode preference in localStorage, default to enabled if not set
         let isDarkModeEnabled = localStorage.getItem('darkMode');
@@ -1985,9 +2200,11 @@ function updatePaginationControls() {
     window.onload = function () {
         setupPage();
         hideProgress(); // Ensure the progress indicator is hidden on page load
+        if (window.location.protocol !== 'http:' && window.location.protocol !== 'https:') {
+            alert('AllTalk TTS Generator is not designed to be run from your disk folder. Please run it from the correct URL e.g. http://127.0.0.1:7851/static/tts_chunk_player/tts_chunk_player.html');
+        }
     };
 
-
 </script>
 </body>
 

system/tts_srt/tts_srt.py

@@ -0,0 +1,77 @@
+import json
+import argparse
+import soundfile as sf
+from pathlib import Path
+
+parser = argparse.ArgumentParser(description="Compare TTS output with the original text using detailed comparison.")
+parser.add_argument("--ttslistpath", help="Path to the ttsList.json file")
+parser.add_argument("--wavfilespath", help="Path to the wav outputs folder")
+args = parser.parse_args()
+
+if not args.ttslistpath:
+    parser.error("[AllTalk TTSSRT] Please specify the path to the ttslist.json file using the --ttslistpath argument.")
+if not args.wavfilespath:
+    parser.error("[AllTalk TTSSRT] Please specify the path to the wavs output folder file using the --wavfilespath argument.")
+
+json_file_path = Path(args.ttslistpath).resolve()
+audio_files_base_path = Path(args.wavfilespath).resolve()
+
+print(f"[AllTalk TTSSRT]")
+print("[AllTalk TTSSRT] \033[92mStarting SRT Subtitle generation:\033[0m")
+print("[AllTalk TTSSRT]")
+print("[AllTalk TTSSRT] \033[94mJSON file  :\033[92m", json_file_path, "\033[0m")
+print("[AllTalk TTSSRT] \033[94mWAV files  :\033[92m", audio_files_base_path, "\033[0m")
+print("[AllTalk TTSSRT]")
+
+# Function to convert time in seconds to SRT time format
+def format_srt_time(seconds):
+    hours = int(seconds // 3600)
+    minutes = int((seconds % 3600) // 60)
+    seconds = seconds % 60
+    milliseconds = int((seconds - int(seconds)) * 1000)
+    return f"{hours:02}:{minutes:02}:{int(seconds):02},{milliseconds:03}"
+
+# Load the JSON data
+with open(json_file_path, 'r') as json_file:
+    data = json.load(json_file)
+
+# Initialize start time and SRT lines
+start_time = 0.0
+srt_lines = []
+
+# Loop through the JSON data to build SRT entries
+for entry in data:
+    file_name = Path(entry['fileUrl']).name  # Get the file name from the URL
+    file_path = audio_files_base_path / file_name
+    print(f"[AllTalk TTSSRT] Processing file: {file_name}")
+
+    if not file_path.exists():
+        print(f"[AllTalk TTSSRT] Audio file does not exist: {file_path}")
+        continue
+
+    try:
+        # Use soundfile to read the duration of the WAV file
+        with sf.SoundFile(str(file_path), 'r') as f:
+            duration = len(f) / f.samplerate
+            end_time = start_time + duration
+            
+            # Format the SRT entry
+            start_timestamp = format_srt_time(start_time)
+            end_timestamp = format_srt_time(end_time)
+            srt_lines.append(f"{len(srt_lines)+1}\n{start_timestamp} --> {end_timestamp}\n{entry['text']}\n")
+            
+            # Update the start time for the next entry
+            start_time = end_time
+    except RuntimeError as e:
+        print(f"[AllTalk TTSSRT] Error reading {file_path}: {e}")
+
+# Join all SRT lines into the final SRT content
+srt_content = "\n".join(srt_lines)
+
+# Save the SRT content to a file
+srt_file_path = audio_files_base_path / "subtitles.srt"
+with open(srt_file_path, "w") as srt_file:
+    srt_file.write(srt_content)
+
+print("[AllTalk TTSSRT]")
+print(f"[AllTalk TTSSRT] \033[94mSRT file created at \033[0m{srt_file_path}")

tts_server.py

@@ -19,6 +19,7 @@ from fastapi import (
     Response,
     Depends,
     HTTPException,
+    Query,
 )
 from fastapi.responses import JSONResponse, HTMLResponse, RedirectResponse, FileResponse, StreamingResponse
 from fastapi.middleware.cors import CORSMiddleware
@@ -34,7 +35,7 @@ this_dir = Path(__file__).parent.resolve()
 # STARTUP VARIABLE - Set "device" to cuda if exists, otherwise cpu
 device = "cuda" if torch.cuda.is_available() else "cpu"
 # STARTUP VARIABLE - Import languges file for Gradio to be able to display them in the interface
-with open(this_dir / "languages.json", encoding="utf8") as f:
+with open(this_dir / "system" / "config" / "languages.json", encoding="utf8") as f:
     languages = json.load(f)
 # Base setting for a possible FineTuned model existing and the loader being available
 tts_method_xtts_ft = False
@@ -57,6 +58,10 @@ params = load_config(configfile_path)
 # check someone hasnt enabled lowvram on a system thats not cuda enabled
 params["low_vram"] = "false" if not torch.cuda.is_available() else params["low_vram"]
 
+# Load values for temperature and repetition_penalty
+temperature = params["local_temperature"]
+repetition_penalty = params["local_repetition_penalty"]
+
 # Define the path to the JSON file
 config_file_path = this_dir / "modeldownload.json"
 
@@ -473,16 +478,16 @@ async def deepspeed(request: Request, new_deepspeed_value: bool):
 ########################
 
 # TTS VOICE GENERATION METHODS (called from voice_preview and output_modifer)
-async def generate_audio(text, voice, language, output_file, streaming=False):
+async def generate_audio(text, voice, language, temperature, repetition_penalty, output_file, streaming=False):
     # Get the async generator from the internal function
-    response = generate_audio_internal(text, voice, language, output_file, streaming)
+    response = generate_audio_internal(text, voice, language, temperature, repetition_penalty, output_file, streaming)
     # If streaming, then return the generator as-is, otherwise just exhaust it and return
     if streaming:
         return response
     async for _ in response:
         pass
     
-async def generate_audio_internal(text, voice, language, output_file, streaming):
+async def generate_audio_internal(text, voice, language, temperature, repetition_penalty, output_file, streaming):
     global model
     if params["low_vram"] and device == "cpu":
         await switch_device()
@@ -504,9 +509,9 @@ async def generate_audio_internal(text, voice, language, output_file, streaming)
             "language": language,
             "gpt_cond_latent": gpt_cond_latent,
             "speaker_embedding": speaker_embedding,
-            "temperature": float(params["local_temperature"]),
+            "temperature": float(temperature),
             "length_penalty": float(model.config.length_penalty),
-            "repetition_penalty": float(params["local_repetition_penalty"]),
+            "repetition_penalty": float(repetition_penalty),
             "top_k": int(model.config.top_k),
             "top_p": float(model.config.top_p),
             "enable_text_splitting": True
@@ -560,9 +565,9 @@ async def generate_audio_internal(text, voice, language, output_file, streaming)
             file_path=output_file,
             speaker_wav=[f"{this_dir}/voices/{voice}"],
             language=language,
-            temperature=float(params["local_temperature"]),
+            temperature=temperature,
             length_penalty=model.config.length_penalty,
-            repetition_penalty=float(params["local_repetition_penalty"]),
+            repetition_penalty=repetition_penalty,
             top_k=model.config.top_k,
             top_p=model.config.top_p,
         )
@@ -602,10 +607,12 @@ async def generate(request: Request):
         text = data["text"]
         voice = data["voice"]
         language = data["language"]
+        temperature = data["temperature"]
+        repetition_penalty = data["repetition_penalty"]
         output_file = data["output_file"]
         streaming = False
         # Generation logic
-        response = await generate_audio(text, voice, language, output_file, streaming)
+        response = await generate_audio(text, voice, language, temperature, repetition_penalty, output_file, streaming)
         if streaming:
             return StreamingResponse(response, media_type="audio/wav")
         return JSONResponse(
@@ -632,7 +639,7 @@ def list_files(directory):
 #############################
 
 # Create an instance of Jinja2Templates for rendering HTML templates
-templates = Jinja2Templates(directory=this_dir / "templates")
+templates = Jinja2Templates(directory=this_dir / "system")
 
 # Create a dependency to get the current JSON data
 def get_json_data():
@@ -647,7 +654,7 @@ async def get_settings(request: Request):
     wav_files = list_files(this_dir / "voices")
     # Render the template with the current JSON data and list of WAV files
     return templates.TemplateResponse(
-        "generate_form.html",
+        "/at_admin/at_settings.html",
         {
             "request": request,
             "data": get_json_data(),
@@ -657,7 +664,7 @@ async def get_settings(request: Request):
     )
 
 # Define an endpoint to serve static files
-app.mount("/static", StaticFiles(directory=str(this_dir / "templates")), name="static")
+app.mount("/static", StaticFiles(directory=str(this_dir / "system")), name="static")
 
 @app.post("/update-settings")
 async def update_settings(
@@ -722,7 +729,7 @@ async def update_settings(
 async def tts_demo_request_streaming(text: str, voice: str, language: str, output_file: str):
     try:
         output_file_path = this_dir / "outputs" / output_file
-        stream = await generate_audio(text, voice, language, output_file_path, streaming=True)
+        stream = await generate_audio(text, voice, language, temperature, repetition_penalty, output_file_path, streaming=True)
         return StreamingResponse(stream, media_type="audio/wav")
     except Exception as e:
         print(f"An error occurred: {e}")
@@ -732,7 +739,7 @@ async def tts_demo_request_streaming(text: str, voice: str, language: str, outpu
 async def tts_demo_request(request: Request, text: str = Form(...), voice: str = Form(...), language: str = Form(...), output_file: str = Form(...)):
     try:
         output_file_path = this_dir / "outputs" / output_file
-        await generate_audio(text, voice, language, output_file_path, streaming=False)
+        await generate_audio(text, voice, language, temperature, repetition_penalty, output_file_path, streaming=False)
         return JSONResponse(content={"output_file_path": str(output_file)}, status_code=200)
     except Exception as e:
         print(f"An error occurred: {e}")
@@ -794,7 +801,7 @@ async def preview_voice(request: Request, voice: str = Form(...)):
 
         # Generate the audio
         output_file_path = this_dir / "outputs" / f"{output_file_name}.wav"
-        await generate_audio(text, voice, language, output_file_path, streaming=False)
+        await generate_audio(text, voice, language, temperature, repetition_penalty, output_file_path, streaming=False)
 
         # Generate the URL
         output_file_url = f'http://{params["ip_address"]}:{params["port_number"]}/audio/{output_file_name}.wav'
@@ -831,7 +838,7 @@ import hashlib
 async def tts_generate_streaming(text: str, voice: str, language: str, output_file: str):
     try:
         output_file_path = this_dir / "outputs" / output_file
-        stream = await generate_audio(text, voice, language, output_file_path, streaming=True)
+        stream = await generate_audio(text, voice, language, temperature, repetition_penalty, output_file_path, streaming=True)
         return StreamingResponse(stream, media_type="audio/wav")
     except Exception as e:
         print(f"An error occurred: {e}")
@@ -841,7 +848,7 @@ async def tts_generate_streaming(text: str, voice: str, language: str, output_fi
 async def tts_generate_streaming(request: Request, text: str = Form(...), voice: str = Form(...), language: str = Form(...), output_file: str = Form(...)):
     try:
         output_file_path = this_dir / "outputs" / output_file
-        await generate_audio(text, voice, language, output_file_path, streaming=False)
+        await generate_audio(text, voice, language, temperature, repetition_penalty, output_file_path, streaming=False)
         return JSONResponse(content={"output_file_path": str(output_file)}, status_code=200)
     except Exception as e:
         print(f"An error occurred: {e}")
@@ -864,7 +871,7 @@ except OSError:
         print(f"[{params['branding']}Startup] \033[91mInfo\033[0m On Linux, you can use the following command to install PortAudio:")
         print(f"[{params['branding']}Startup] \033[91mInfo\033[0m sudo apt-get install portaudio19-dev")
 
-from typing import Union, Dict
+from typing import Union, Dict, List
 from pydantic import BaseModel, ValidationError, Field
 
 def play_audio(file_path, volume):
@@ -877,7 +884,7 @@ class Request(BaseModel):
     pass
 
 class JSONInput(BaseModel):
-    text_input: str = Field(..., max_length=1000, description="text_input needs to be 1000 characters or less.")
+    text_input: str = Field(..., max_length=2000, description="text_input needs to be 2000 characters or less.")
     text_filtering: str = Field(..., pattern="^(none|standard|html)$", description="text_filtering needs to be 'none', 'standard' or 'html'.")
     character_voice_gen: str = Field(..., pattern="^.*\.wav$", description="character_voice_gen needs to be the name of a wav file e.g. mysample.wav.")
     narrator_enabled: bool = Field(..., description="narrator_enabled needs to be true or false.")
@@ -1054,7 +1061,7 @@ async def tts_generate(
                 cleaned_part = re.sub(r'\n+', ' ', cleaned_part)
                 output_file = this_dir / "outputs" / f"{output_file_name}_{uuid.uuid4()}_{int(time.time())}.wav"
                 output_file_str = output_file.as_posix()
-                response = await generate_audio(cleaned_part, voice_to_use, language, output_file_str, streaming)
+                response = await generate_audio(cleaned_part, voice_to_use, language,temperature, repetition_penalty, output_file_str, streaming)
                 audio_path = output_file_str
                 audio_files_all_paragraphs.append(audio_path)
             # Combine audio files across paragraphs
@@ -1091,7 +1098,7 @@ async def tts_generate(
                 cleaned_string = re.sub(r'\n+', ' ', cleaned_string)
             else:
                 cleaned_string = text_input
-            response = await generate_audio(cleaned_string, character_voice_gen, language, output_file_path, streaming)
+            response = await generate_audio(cleaned_string, character_voice_gen, language, temperature, repetition_penalty, output_file_path, streaming)
         if sounddevice_installed == False or streaming == True:
             autoplay = False
         if autoplay:
@@ -1139,7 +1146,69 @@ def get_current_settings():
 #### Word Add-in Sharing ####
 #############################
 # Mount the static files from the 'word_addin' directory
-app.mount("/api/word_addin", StaticFiles(directory=os.path.join(this_dir / 'templates' / 'word_addin')), name="word_addin")
+app.mount("/api/word_addin", StaticFiles(directory=os.path.join(this_dir / 'system' / 'word_addin')), name="word_addin")
+
+#############################################
+#### TTS Generator Comparision Endpoints ####
+#############################################
+import subprocess
+import aiofiles
+
+class TTSItem(BaseModel):
+    id: int
+    fileUrl: str
+    text: str
+    characterVoice: str
+    language: str
+
+class TTSData(BaseModel):
+    ttsList: List[TTSItem]
+
+@app.post("/api/save-tts-data")
+async def save_tts_data(tts_data: List[TTSItem]):
+    # Convert the list of Pydantic models to a list of dictionaries
+    tts_data_list = [item.dict() for item in tts_data]
+    # Serialize the list of dictionaries to a JSON string
+    tts_data_json = json.dumps(tts_data_list, indent=4)
+    async with aiofiles.open(this_dir / "outputs" / "ttsList.json", 'w') as f:
+        await f.write(tts_data_json)
+    return {"message": "Data saved successfully"}
+
+import sys
+
+@app.get("/api/trigger-analysis")
+async def trigger_analysis(threshold: int = Query(default=98)):
+    venv_path = sys.prefix
+    env = os.environ.copy()
+    env["PATH"] = os.path.join(venv_path, "bin") + ":" + env["PATH"]
+    ttslist_path = this_dir / "outputs" / "ttsList.json"
+    wavfile_path = this_dir / "outputs"
+    subprocess.run(["python", "tts_diff.py", f"--threshold={threshold}", f"--ttslistpath={ttslist_path}", f"--wavfilespath={wavfile_path}"], cwd=this_dir / "system" / "tts_diff", env=env)
+    # Read the analysis summary
+    try:
+        with open(this_dir / "outputs" / "analysis_summary.json", "r") as summary_file:
+            summary_data = json.load(summary_file)
+    except FileNotFoundError:
+        summary_data = {"error": "Analysis summary file not found."}
+    return {"message": "Analysis Completed", "summary": summary_data}
+
+#################################################
+#### TTS Generator SRT Subtitiles generation ####
+#################################################
+@app.get("/api/srt-generation")
+async def srt_generation():
+    venv_path = sys.prefix
+    env = os.environ.copy()
+    env["PATH"] = os.path.join(venv_path, "bin") + ":" + env["PATH"]
+    ttslist_path = this_dir / "outputs" / "ttsList.json"
+    wavfile_path = this_dir / "outputs"
+    subprocess.run(["python", "tts_srt.py", f"--ttslistpath={ttslist_path}", f"--wavfilespath={wavfile_path}"], cwd=this_dir / "system" / "tts_srt", env=env)
+    
+    srt_file_path = this_dir / "outputs" / "subtitles.srt"
+    if not srt_file_path.exists():
+        raise HTTPException(status_code=404, detail="Subtitle file not found.")
+    
+    return FileResponse(path=srt_file_path, filename="subtitles.srt", media_type='application/octet-stream')
 
 ###################################################
 #### Webserver Startup & Initial model Loading ####

voices/Clint_Eastwood CC3.txt

@@ -9,4 +9,6 @@ Author	BBC
 This file is licensed under the Creative Commons Attribution 3.0 Unported license.
 https://creativecommons.org/licenses/by/3.0/deed.en
 
-Downsampled to 22050hz, mono and cut shorter
+Downsampled to 22050hz, mono and cut shorter.
+
+This license applies to both the original file ( «Clint_Eastwood CC3.wav» ) and it's derivative, the enhanced file ( «Clint_Eastwood (enhanced)CC3.wav» ).