search

modelscope / FunCodec

FunCodec is a research-oriented toolkit for audio quantization and downstream applications, such as text-to-speech synthesis, music generation et.al.
https://funcodec.github.io/
MIT License
370 stars 30 forks source link

Has Funcodec-based TTS been released yet？ #19

Closed WhuJing closed 8 months ago

WhuJing commented 10 months ago

when would Funcodec-based text-to-speech be relesead？

ZhihaoDU commented 10 months ago

Thanks for you attention. text-to-speech English model, LauraTTS, has already been released, which it trained on LibriTTS. You can find it at egs/LibriTTS/text2speech_laura.