linto-ai / whisper-timestamped

Multilingual Automatic Speech Recognition with word-level timestamps and confidence
GNU Affero General Public License v3.0
2.01k stars 156 forks source link

Verbose option does not support None (no output) #25

Closed kamranjon closed 1 year ago

kamranjon commented 1 year ago

In the original Whisper transcribe implementation providing None to the verbose option suppresses everything, which is explained here. I believe it is this line here, which overrides the None behavior and subsequently there is no option to have no output whatsoever (providing None shows the progress bar and language, providing False also shows the progress Bar and Providing True does not show the progress bar but displays the list of word-level timestamps).

It would be ideal if there was an option to suppress all output from the transcribe method.

Jeronymous commented 1 year ago

Thanks for spotting @kamranjon I fixed that