search

hubertsiuzdak / snac

Multi-Scale Neural Audio Codec (SNAC) compresses audio into discrete codes at a low bitrate
https://hubertsiuzdak.github.io/snac/
MIT License
443 stars 26 forks source link

Is this could be used for audio synthesis? #24

Open MonolithFoundation opened 2 months ago

MonolithFoundation commented 2 months ago

For instance, LLM out produce snac tokens, and decode into audio?

itsliupeng commented 2 months ago

FYI. https://github.com/gpt-omni/mini-omni uses snac codec to generate audio.

MonolithFoundation commented 2 months ago

Thanks for the hint, how about Chinese?

MrWaterZhou commented 6 days ago

https://github.com/MrWaterZhou/viitor-voice

I tried, and it works well :)

MonolithFoundation commented 6 days ago

Woo, does it support Madrian and Japanese?

MrWaterZhou commented 6 days ago

Woo, does it support Madrian and Japanese?

Not yet, but we are working on Madrian and will release it soon.