search

NavodPeiris / speechlib

speechlib is a library that can do speaker diarization, transcription and speaker recognition on an audio file to create transcripts with actual speaker names
MIT License
114 stars 10 forks source link

Any plans to use fater-whisper? #3

Closed dgoryeo closed 5 months ago

dgoryeo commented 5 months ago

Hi, thanks for the good work. SInce there was no placholder for discussion thread I am using this as a feature suggestion: whould you have any plans to use fater-whisper models --for the purpose reducing VRAM and processing time? Thanks.

NavodPeiris commented 5 months ago

Thanks for the suggestion. I may add this in future!

NavodPeiris commented 5 months ago

@dgoryeo added the faster-whisper. works like a charm!