Merge pull request #557 from Paladinium/alltalkbeta_docker_env_variables

Introduce ENV variables for Docker

DOCKER_README.md

@@ -53,7 +53,7 @@ following optional arguments:
     - Example: `docker-start.sh --voices /my/voices/dir`
 - `--rvc-voices` similar to voices, this option lets you pick the folder containing the [RVC models](https://github.com/erew123/alltalk_tts/wiki/RVC-(Retrieval%E2%80%90based-Voice-Conversion)).
     - Example: `docker-start.sh --rvc_voices /my/rvc/voices/dir`
-- `--no-ui` allows you to not expose port 7852 for the gradio interface and sets `launch_gradio` to `false`.
+- `--no-ui` allows you to not expose port 7852 for the gradio interface and sets `launch_gradio` and `gradio_interface` to `false`.
 - `--with-multi-engine-manager` enables the use of the [multi engine manager (MEM)](https://github.com/erew123/alltalk_tts/wiki/Multi-Engine-Manager)
     which allows for more parallel requests. By default, one TTS engine is started. Optionally, you can pass the
     file path of a JSON file which can be a subset of `mem_config.json` with more fine-grained configuration options.
@@ -66,8 +66,27 @@ following optional arguments:
 - Since the above commands only address the most important options, you might pass additional arbitrary docker arguments
   to the `docker-start.sh`.
 
-Of course, like any other Docker image, you can also directly use `docker run` directly and use `docker-start.sh` 
-as an inspiration.
+Of course, like for any other Docker image, you can also directly use `docker run` directly and use 
+[docker-start.sh](https://github.com/erew123/alltalk_tts/blob/alltalkbeta/docker-start.sh) as an inspiration.
+
+### Using environment variables for setting config values
+Instead of using config JSON files, there are config settings that can be directly set using environment (ENV) variables 
+when starting the application. This is also what is used internally in `docker-start.sh`.
+
+Since `docker-start.sh` allows to pass arbitrary params, it can also be used to set env variables, e.g:
+
+```
+./docker-start.sh -e ALLTALK_BRANDING="My Brand " -e ALLTALK_API_PORT_NUMBER=9876
+```
+
+or when using Docker directly:
+```
+docker run -e ALLTALK_BRANDING="My Brand " -e ALLTALK_API_PORT_NUMBER=9876 ...
+```
+
+Consult [docker_default_confignew.json](https://github.com/erew123/alltalk_tts/blob/alltalkbeta/docker_default_confignew.json) 
+and [docker_default_mem_config.json](https://github.com/erew123/alltalk_tts/blob/alltalkbeta/docker_default_mem_config.json) 
+for a list of all available env variables.
 
 
 ## Building your own images (advanced)

Dockerfile

@@ -13,8 +13,18 @@ ARG ALLTALK_DIR=/opt/alltalk
 
 ENV GRADIO_SERVER_NAME="0.0.0.0"
 
-ENV ENABLE_MULTI_ENGINE_MANAGER=false
-ENV WITH_UI=true
+# Switch for using the multi engine manager:
+ENV ALLTALK_ENABLE_MULTI_ENGINE_MANAGER=false
+
+# Default settings used in docker_default_confignew.json:
+ENV ALLTALK_DELETE_OUTPUT_WAVS="1"
+ENV ALLTALK_GRADIO_INTERFACE=true
+ENV ALLTALK_RVC_ENABLED=true
+ENV ALLTALK_RVC_F0METHOD="rmvpe"
+
+# Default settings used in docker_default_mem_config.json:
+ENV ALLTALK_MEM_AUTO_START_ENGINES=1
+ENV ALLTALK_MEM_MAX_INSTANCES=1
 
 WORKDIR ${ALLTALK_DIR}
 
@@ -118,21 +128,28 @@ RUN <<EOR
 #!/usr/bin/env bash
 source ~/.bashrc
 
-# Enabling or disabling UI:
-jq ".launch_gradio = \$WITH_UI" docker_default_config.json > docker_default_config.json.tmp
-mv docker_default_config.json.tmp docker_default_config.json
+replace_env_vars() {
+  echo "{}" > /tmp/empty.json
+  jq "\$( cat \$1 ) | del(.. | select(. == null))" /tmp/empty.json > \$1.tmp
+  mv \$1.tmp \$1
+  rm -f /tmp/empty.json
+}
 
-# Merging config from docker_confignew.json into confignew.json:
-jq -s '.[0] * .[1] * .[2]' confignew.json docker_default_config.json docker_confignew.json  > confignew.json.tmp
-mv confignew.json.tmp confignew.json
+merge_json_files() {
+  # Merging JSON config with docker default values followed by values set on startup:
+  jq -s '.[0] * .[1] * .[2]' \$1 docker_default_\$1 docker_\$1  > \$1.tmp
+  mv \$1.tmp \$1
+}
+
+replace_env_vars docker_default_confignew.json
+merge_json_files confignew.json
 
 conda activate alltalk
 
-if [ "\$ENABLE_MULTI_ENGINE_MANAGER" = true ] ; then
+if [ "\$ALLTALK_ENABLE_MULTI_ENGINE_MANAGER" = true ] ; then
   echo "Starting alltalk using multi engine manager"
-  # Merging config from docker_mem_config.json into mem_config.json:
-  jq -s '.[0] * .[1] * .[2]' mem_config.json docker_default_mem_config.json docker_mem_config.json  > mem_config.json.tmp
-  mv mem_config.json.tmp mem_config.json
+  replace_env_vars docker_default_mem_config.json
+  merge_json_files mem_config.json
   python tts_mem.py
 else
   echo "Starting alltalk"

docker-start.sh

@@ -3,8 +3,8 @@
 SCRIPT_DIR=$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )
 . ${SCRIPT_DIR=}/docker/variables.sh
 
-WITH_UI=true
-ENABLE_MULTI_ENGINE_MANAGER=false
+ALLTALK_GRADIO_INTERFACE=true
+ALLTALK_ENABLE_MULTI_ENGINE_MANAGER=false
 MULTI_ENGINE_MANAGER_CONFIG=
 DOCKER_TAG=latest-xtts
 DOCKER_REPOSITORY=erew123
@@ -26,10 +26,10 @@ while [ "$#" -gt 0 ]; do
       shift
       ;;
     --no-ui)
-      WITH_UI=false
+      ALLTALK_GRADIO_INTERFACE=false
       ;;
     --with-multi-engine-manager)
-      ENABLE_MULTI_ENGINE_MANAGER=true
+      ALLTALK_ENABLE_MULTI_ENGINE_MANAGER=true
       if [ -n "$2" ] && ! [[ $2 =~ ^--.* ]]; then
         MULTI_ENGINE_MANAGER_CONFIG=$2
         shift
@@ -74,8 +74,8 @@ if [[ -n $RVC_VOICES ]]; then
   DOCKER_ARGS+=( -v ${RVC_VOICES}:${ALLTALK_DIR}/models/rvc_voices )
 fi
 
-if [ "$WITH_UI" = true ] ; then
-    if [ "$ENABLE_MULTI_ENGINE_MANAGER" = true ] ; then
+if [ "ALLTALK_GRADIO_INTERFACE" = true ] ; then
+    if [ "$ALLTALK_ENABLE_MULTI_ENGINE_MANAGER" = true ] ; then
       DOCKER_ARGS+=( -p 7500:7500 )
     else
       DOCKER_ARGS+=( -p 7852:7852 )
@@ -87,8 +87,10 @@ if [[ -n $MULTI_ENGINE_MANAGER_CONFIG ]]; then
 fi
 
 # Pass env variables:
-DOCKER_ARGS+=( -e ENABLE_MULTI_ENGINE_MANAGER=${ENABLE_MULTI_ENGINE_MANAGER} )
-DOCKER_ARGS+=( -e WITH_UI=${WITH_UI} )
+DOCKER_ARGS+=( -e ALLTALK_ENABLE_MULTI_ENGINE_MANAGER=${ALLTALK_ENABLE_MULTI_ENGINE_MANAGER} )
+DOCKER_ARGS+=( -e ALLTALK_GRADIO_INTERFACE=${ALLTALK_GRADIO_INTERFACE} )
+DOCKER_ARGS+=( -e ALLTALK_LAUNCH_GRADIO=${ALLTALK_GRADIO_INTERFACE} )
+
 
 docker run \
   --rm \

docker_default_config.json

@@ -1,7 +0,0 @@
-{
-  "delete_output_wavs": "1",
-  "rvc_settings": {
-    "rvc_enabled": true,
-    "f0method": "rmvpe"
-  }
-}

docker_default_confignew.json

@@ -0,0 +1,40 @@
+{
+  "branding": $ENV.ALLTALK_BRANDING,
+  "delete_output_wavs": $ENV.ALLTALK_DELETE_OUTPUT_WAVS,
+  "gradio_interface": $ENV.ALLTALK_GRADIO_INTERFACE,
+  "output_folder": $ENV.ALLTALK_OUTPUT_FOLDER,
+  "gradio_port_number": $ENV.ALLTALK_GRADIO_PORT_NUMBER,
+  "launch_gradio": $ENV.ALLTALK_LAUNCH_GRADIO,
+  "transcode_audio_format": $ENV.ALLTALK_TRANSCODE_AUDIO_FORMAT,
+  "rvc_settings": {
+    "rvc_enabled": $ENV.ALLTALK_RVC_ENABLED,
+    "rvc_char_model_file": $ENV.ALLTALK_RVC_CHAR_MODEL_FILE,
+    "rvc_narr_model_file": $ENV.ALLTALK_RVC_NARR_MODEL_FILE,
+    "split_audio": $ENV.ALLTALK_RVC_SPLIT_AUDIO,
+    "autotune": $ENV.ALLTALK_RVC_AUTOTUNE,
+    "pitch": $ENV.ALLTALK_RVC_PITCH,
+    "filter_radius": $ENV.ALLTALK_RVC_FILTER_RADIUS,
+    "index_rate": $ENV.ALLTALK_RVC_INDEX_RATE,
+    "rms_mix_rate": $ENV.ALLTALK_RVC_RMS_MIX_RATE,
+    "protect": $ENV.ALLTALK_RVC_PROTECT,
+    "hop_length": $ENV.ALLTALK_RVC_HOP_LENGTH,
+    "f0method": $ENV.ALLTALK_RVC_F0METHOD,
+    "embedder_model": $ENV.ALLTALK_RVC_EMBEDDER_MODEL,
+    "training_data_size": $ENV.ALLTALK_RVC_TRAINING_DATA_SIZE
+  },
+  "api_def": {
+    "api_port_number": $ENV.ALLTALK_API_PORT_NUMBER,
+    "api_allowed_filter": $ENV.ALLTALK_API_ALLOWED_FILTER,
+    "api_length_stripping": $ENV.ALLTALK_API_LENGTH_STRIPPING,
+    "api_max_characters": $ENV.ALLTALK_API_MAX_CHARACTERS,
+    "api_use_legacy_api": $ENV.ALLTALK_API_USE_LEGACY_API,
+    "api_text_filtering": $ENV.ALLTALK_API_TEXT_FILTERING,
+    "api_narrator_enabled": $ENV.ALLTALK_API_NARRATOR_ENABLED,
+    "api_text_not_inside": $ENV.ALLTALK_API_TEXT_NOT_INSIDE,
+    "api_language": $ENV.ALLTALK_API_LANGUAGE,
+    "api_output_file_name": $ENV.ALLTALK_API_OUTPUT_FILE_NAME,
+    "api_output_file_timestamp": $ENV.ALLTALK_API_OUTPUT_FILE_TIMESTAMP,
+    "api_autoplay": $ENV.ALLTALK_API_AUTOPLAY,
+    "api_autoplay_volume": $ENV.ALLTALK_API_AUTOPLAY_VOLUME
+  }
+}

docker_default_mem_config.json

@@ -1,3 +1,18 @@
 {
-  "auto_start_engines": 1
+  "base_port": $ENV.ALLTALK_MEM_BASE_PORT,
+  "api_server_port": $ENV.ALLTALK_MEM_API_SERVER_PORT,
+  "auto_start_engines": $ENV.ALLTALK_MEM_AUTO_START_ENGINES,
+  "max_instances": $ENV.ALLTALK_MEM_MAX_INSTANCES,
+  "gradio_interface_port": $ENV.ALLTALK_MEM_GRADIO_INTERFACE_PORT,
+  "max_retries": $ENV.ALLTALK_MEM_MAX_RETRIES,
+  "initial_wait": $ENV.ALLTALK_MEM_INITIAL_WAIT,
+  "backoff_factor": $ENV.ALLTALK_MEM_BACKOFF_FACTOR,
+  "debug_mode": $ENV.ALLTALK_MEM_DEBUG_MODE,
+  "max_queue_time": $ENV.ALLTALK_MEM_MAX_QUEUE_TIME,
+  "queue_check_interval": $ENV.ALLTALK_MEM_QUEUE_CHECK_INTERVAL,
+  "tts_request_timeout": $ENV.ALLTALK_MEM_TTS_REQUEST_TIMEOUT,
+  "text_length_factor": $ENV.ALLTALK_MEM_TEXT_LENGTH_FACTOR,
+  "concurrent_request_factor": $ENV.ALLTALK_MEM_CONCURRENT_REQUEST_FACTOR,
+  "diminishing_factor": $ENV.ALLTALK_MEM_DIMINISHING_FACTOR,
+  "queue_position_factor": $ENV.ALLTALK_MEM_QUEUE_POSITION_FACTOR
 }

tts_mem.py

@@ -55,7 +55,7 @@ from collections import deque
 from datetime import datetime, timedelta
 from requests.exceptions import RequestException
 from werkzeug.serving import make_server, WSGIRequestHandler
-from config import AlltalkMultiEngineManagerConfig
+from config import AlltalkMultiEngineManagerConfig, AlltalkConfig
 from flask_cors import CORS, cross_origin
 from flask import Flask, request, jsonify, send_from_directory, send_file
 
@@ -76,7 +76,6 @@ should_exit = threading.Event()
 # Global variable to hold the monitor thread
 stop_monitor = threading.Event()
 monitor = None
-WITH_UI = os.getenv("WITH_UI", True).lower() == 'true'
 
 flask_app = Flask(__name__)
 CORS(flask_app)
@@ -84,6 +83,7 @@ flask_app.config['OUTPUT_FOLDER'] = str(Path(os.getcwd()) / 'outputs')
 
 # Load configuration at the start of the script
 config = AlltalkMultiEngineManagerConfig.get_instance()
+USE_GRADIO = AlltalkConfig.get_instance().gradio_interface
 
 def signal_handler(signum, frame):
     print("[AllTalk MEM] Interrupt received, shutting down...")
@@ -125,7 +125,7 @@ print(f"[AllTalk MEM] \033[93m     MEM is not intended for production use and\03
 print(f"[AllTalk MEM] \033[93m      there is NO support being offered on MEM\033[00m")
 print(f"[AllTalk MEM]")
 print(f"[AllTalk MEM] \033[94mAPI/Queue   :\033[00m \033[92mhttp://127.0.0.1:{config.api_server_port}/api/tts-generate\033[00m")
-if WITH_UI:
+if USE_GRADIO:
     print(f"[AllTalk MEM] \033[94mGradio Light:\033[00m \033[92mhttp://127.0.0.1:{config.gradio_interface_port}\033[00m")
     print(f"[AllTalk MEM] \033[94mGradio Dark :\033[00m \033[92mhttp://127.0.0.1:{config.gradio_interface_port}?__theme=dark\033[00m")
 print(f"[AllTalk MEM]")
@@ -1492,7 +1492,7 @@ def auto_start_engines():
 # Main execution
 if __name__ == "__main__":
     signal.signal(signal.SIGINT, signal_handler)
-    if WITH_UI:
+    if USE_GRADIO:
         print("[AllTalk MEM] Starting gradio...")
         interface = create_gradio_interface()
         # Start the Gradio interface