An extension to use GPT-SoVITS in text generation webui, it uses the v2 version of the API

If you not familiar with this TTS, just type the relative path from the GPT SoVits or the absolute path of the audio file, the prompt text is the transcription of the audio and isn't required. Filling the languages will improve the quality!

Install from Session tab of text generation webui or Clone this repository into the extension folder

Follow the instruction of GPT SoVits or download the package in realease (I recommend the v2) In the config.py of the GPT SoVits fill with you model's paths, it's possible to change model through the API but I'm not doing that

And finally, to launch the API server run runtime\python.exe .\api_v2.py -a 0.0.0.0 if you downloaded the package.