Spaces:

TTS-AGI

/

TTS-Arena

like 501

Running on CPU Upgrade

TTS name - VoiceCraft
Authors - (not me) Puyuan Peng, Po-Yao Huang, Daniel Li, Abdelrahman Mohamed, David Harwath
Paper - https://jasonppy.github.io/assets/pdfs/VoiceCraft.pdf
Github - https://github.com/jasonppy/VoiceCraft
Demo - https://jasonppy.github.io/VoiceCraft_web/
Huggingface Link - https://huggingface.co/pyp1/VoiceCraft/tree/main
Model License: Coqui Public Model License 1.0.0

I've recently heard about this model, and the quality seems to be very good. There are two sizes: 330M and 830M. I'm not sure about the model license though, since Coqui doesn't really exist anymore as a company. It should be good for non-commercial use.

On another note, it may also be a good idea to add a feature for a user to upload a snippet to clone.