search

thomasmol / cog-whisper-diarization

Cog implementation of transcribing + diarization pipeline with Whisper & Pyannote
https://replicate.com/thomasmol/whisper-diarization
165 stars 51 forks source link

[FEATURE REQUEST] Add Batched faster-whisper #18

Closed NickNaskida closed 1 month ago

NickNaskida commented 1 month ago

Hey, thanks a lot for your work here! Have you considered using Batched faster-whisper?

As per docs:

Batched version improves the speed upto 10-12x compared to openAI implementation and 3-4x compared to the sequential faster_whisper version. It works by transcribing semantically meaningful audio chunks as batches leading to faster inference.

I am down to create a pull request adding this. Let me know.

thomasmol commented 1 month ago

Yes I have seen it! its awesome. however it is not released yet officially as there are a few issues, e.g. https://github.com/SYSTRAN/faster-whisper/pull/940 so i'll wait until an official new release to add this.

NickNaskida commented 1 month ago

Cool, I'll deploy my own until then

NickNaskida commented 1 month ago

If anyone needs it, it is here: https://replicate.com/nicknaskida/whisper-diarization

Closing this issue for now.