Spaces:
Paused
Apply for a GPU community grant: Academic project
Hi HF team! This is the official demo for GibbsTTS (arxiv.org/abs/2605.09386), a zero-shot voice-cloning TTS model based on
metric-induced discrete flow matching with a kinetic-optimal scheduler and finite-step CTMC moment correction. The model supports
English and Mandarin Chinese voice cloning from a few seconds of reference audio. We'd love to share this with the community via ZeroGPU
so anyone can try it. The Space uses ~5 GB of GPU memory peak. Thanks!
Paper: arxiv.org/abs/2605.09386
Code: github.com/ydqmkkx/GibbsTTS
Model: huggingface.co/ydqmkkx/GibbsTTS
Hi @caiyiyi1998 ,
Thanks for the request. We noticed that another author from the same paper also submitted a request for the same demo. https://huggingface.co/spaces/ydqmkkx/GibbsTTS/discussions/1
Could you coordinate with your co-author and let us know which request should remain active? Once confirmed, we'll assign ZeroGPU to the Space.
Docs for ZeroGPU: https://huggingface.co/docs/hub/spaces-zerogpu
Hi HF team, Thanks a lot for catching this and for getting back to us.
We've coordinated within the team — please proceed with the request on ydqmkkx/GibbsTTS, that one will be the official demo going forward. We'll pause this Space (caiyiyi1998/gibbstts-demo) on our side so it doesn't stay active.
Thanks again 🙏
Thanks for confirming! I've assigned ZeroGPU to the other Space, as requested.