Open-source text-to-speech
GPT-SoVITS
A voice-cloning and text-to-speech toolkit for zero-shot inference and few-shot speaker adaptation across languages.
- Voice cloning
- Few-shot adaptation
- Cross-language speech
Version & availability
- Project release
- V5
- Embedded demo
- V2 Pro Plus
- Languages
- Chinese, English, Japanese, Korean and Cantonese
The upstream project includes V5, but its recommended public Space currently runs V2 Pro Plus. The embedded demo does not provide V5 inference.
Last checked:
Try GPT-SoVITS
Demo: V2 Pro Plus · Maintainer demo linked by the project
Hosted on Hugging Face. GPU queues, sign-in or usage limits may apply. If the demo does not load here, open it directly using the link above.
License & model terms
Code: MIT. Check the terms of any pretrained weights and reference voices you use.
Read the project documentation ↗