* adds langdetect as requirement for colab, standalone and textgen
* adds "auto" to the language dropdown in the Advanced Engine/Model Settings panel
* replace the hardcoded "en" by "auto" when called by the OpenAI compatible Speech API
**Command-Line Arguments**:
- Accepts a `--tts_model` argument to bypass the interactive menu and directly set up a specific TTS engine
on the first time start-up:
- `piper`: Sets up the Piper TTS engine.
- `vits`: Sets up the VITS TTS engine.
- `xtts`: Sets up the XTTS TTS engine.
- `none`: Skips model setup entirely.
Cleaned up the start-up screen & added mini diags
Added a config file updater that can be used to add/remove things from the confignew.json file if ever needed.
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.
AllTalk will now strip any image from TTS and re-add the image data after TTS.
This is so that when used with Stable Diffusion Image generation within Text-generation-webui, AllTalk isnt trying to generate the Image data as TTS.
ekoc