search

MahmoudAshraf97 / whisper-diarization

Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper
BSD 2-Clause "Simplified" License
3.28k stars 272 forks source link

add device arg for cpu support, Silicon M1 fix, display execution time #40

Closed smxsm closed 1 year ago

smxsm commented 1 year ago

Hi, thanks for your awesome project! Works great and is the easiest to setup and use for Whisper and diarization that I could find so far 😃 I have no GPU though, so I've added a command line param "--device" to change from cuda to cpu if needed. I also had an error on my M1 Silicon Macbook Pro with "num_workers = 1", so I've added a condition there to change it to 0 ... Not sure if you want to integrate that into your scripts, but here is a PR just in case 😉

smxsm commented 1 year ago

Hi, sure, I'll try to incorporate those changes ... so you think I should remove the execution time calc from the scripts?

MahmoudAshraf97 commented 1 year ago

I think so yes, maybe in the future I'll add a nice progress bar or meaningful terminal progress

On Thu, May 11, 2023, 12:45 PM Stefan Moises @.***> wrote:

Hi, sure, I'll try to incorporate those changes ... so you think I should remove the execution time calc from the scripts?

— Reply to this email directly, view it on GitHub https://github.com/MahmoudAshraf97/whisper-diarization/pull/40#issuecomment-1543682590, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHXHGLGMYGKZ2ZSP6H2AVSLXFSYNPANCNFSM6AAAAAAX4533GE . You are receiving this because you commented.Message ID: @.***>

MahmoudAshraf97 commented 1 year ago

Thanks for your contribution