I create a "half" streaming audio transcriber demo

#28
by zhenxuan - opened

Whisper is such a great model and I found it far better to recognizes multiple English accents than other open source models or commerical APIs/SDKs.

Thanks for your work!

I create a "half" streaming transcriber and it's here: https://huggingface.co/spaces/MLSquad-TWCN/near-continuous-whispering

It's lag and user experience is bad even as a demo but I think it's pretty interesting.

Sign up or log in to comment