Mijie
@mijie.eth
The new function on fish.audio looks good. It only takes about 15 seconds of original sound input to simulate human voice. It supports eight major language output, and the model is open source. https://t.co/aVFsUBlnTe
0 reply
0 recast
0 reaction