huggingface / distil-whisper

Distilled variant of Whisper for speech recognition. 6x faster, 50% smaller, within 1% word error rate.
MIT License
3.33k stars 238 forks source link

Streaming Implementation of Distil Whisper #41

Closed souvikqb closed 7 months ago

souvikqb commented 7 months ago

Is it possible to use Distil Whisper for streaming applications?

Can I send small chunks of audio files to a server for processing?

Use Case - I wish to use my mic to send small chunks of audio files to a server and process them quickly to get back the transcripts

sanchit-gandhi commented 7 months ago

Here's a version that runs locally that you can adapt: https://github.com/huggingface/distil-whisper/issues/4#issuecomment-1794494419

sanchit-gandhi commented 7 months ago

Going to close this and encourage the discussion be kept on the original issue! Feel free to comment there if you have any issues!