search

m-bain / whisperX

WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
BSD 2-Clause "Simplified" License
12k stars 1.26k forks source link

Retrieving Word-Level Time Codes with Faster Whisper without Using an Alignment Model #610

Open ImadSidhoum opened 10 months ago

ImadSidhoum commented 10 months ago

I want to have the time codes at the word level, but for the need of inference time optimization, I just want to retrieve the time codes of faster whisper and not use the alignment model. Has anyone done this? Thank you

MahmoudAshraf97 commented 10 months ago

You can do this using faster-whisper directly, but it's not supported in the batched inference here