Spaces:
Running
Running
divide between speakers
#7
by
textToSQL
- opened
In an audio with several speaking people, like an interview, is it possible , while transcribing add a setting to recognize say {speaker1}, {speaker2}, etc.? I can do that with the transcribed text, was wondering if whisper-jax could do that too. Thank you
Hey @textToSQL - this is the task of speaker diarization - there's a discussion on this in the Whisper JAX repo that you might be interested in checking out: https://github.com/sanchit-gandhi/whisper-jax/issues/25
Closing this issue as a duplicate - we can track it on the Whisper JAX repo @textToSQL 🤗
sanchit-gandhi
changed discussion status to
closed