m-bain / whisperX

WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
BSD 2-Clause "Simplified" License
10.32k stars 1.09k forks source link

Is there no more word-level output? #359

Open emremigh opened 1 year ago

emremigh commented 1 year ago

6 months ago I started following the whisperx project and I haven't looked at it for a long time, I think something has changed in the project or something is wrong with me, I did a lot of research and experimentation but I couldn't get word level output in the current version.

I remember that it gives an output file ending with *word.srt next to .srt and .vtt extensions, here is the parameter I used: !whisperx filename.wav --model large-v2 --highlight_words True

can you inform us about the current usage commands?

ppoudd1 commented 1 year ago

I had the same problem