How to implement diarization using Whisper model

[Arasun01]

How do we will find the speaker-level transcripts using whisper. We have a conversation audio file, I would like to find the transcripts per speaker.

[phineas-pta]

search for "diarization" u find other people sharing their code

[Arasun01]

I was following the thread https://towardsdatascience.com/unlock-the-power-of-audio-data-advanced-transcription-and-diarization-with-whisper-whisperx-and-ed9424307281. It does not work for me with the latest version.

[phineas-pta]

ask its author

[Tex2002ans]

How do we will find the speaker-level transcripts using whisper. [...] I would like to find the transcripts per speaker.

A few months ago I answered this in:

• What is the best way to transcribe a conversation? #1191 (comment)

and linked linked to many of the previous "speaker diarization" topics covering this.

[Majdoddin]

@Arasun01 @phineas-pta @Tex2002ans
www.lexicaps.com seamlessly adds diarization to Whispers transcription. No 3rd party packages.
Announcement: #1537
Repo: https://github.com/Majdoddin/lexicaps

[ngtrphuong]

@Majdoddin - The domain you provided is in-accessible

[Arasun01]

Thanks, I will take a look.