Open ZenXir opened 6 months ago
This might be even more relevant now with [distil-large-v3 (already supported with faster-whisper)](Support distil-large-v3 (already supported with faster-whisper)
With multiple servers I suppose an efficient setup could be to to use a server with Large-v2 to perform language detection and transcription of other languages than English, and then use another instance with distil-large-v3 to handle English.
Distil-Whisper is a distilled version of Whisper that is 6 times faster, 49% smaller, and performs within 1% word error rate (WER) on out-of-distribution evaluation sets:
https://github.com/huggingface/distil-whisper.git