speaker diarization | speaker diarization using Whisper | speaker diarization labelling

In this video i have made an effort to explain and demonstrate Speaker diarization using open AI whsiper library & python

In short, Who has spoken what and at what time.

Google Collab link

Hugging Face link

if you have single speaker in audio, please watch below video

[sowmyareddytummuri]

Very informative raghava garu.Good session

[monaligohil]

Hi Raghav, Thanks for the vedio. Could you share the information how can we identify how many speakers are there before assigning any values? Like if some audio is having 2 person in some it's 3. So without any manual efforts how can we just identify how many speakers will be there?

[pohidie]

Hello there! Firstly, thank you. Secondly, I have two questions: 1) Please clarify if this is working and if it will stop working sometime in the future. 2) If I want to transcribe sensitive data, what should be my main concerns, and.. is it even possible to keep it encrypted or secured in some way?

[MonuAgarwal-gvbl]

I have tested the same colab file many times but I am getting the error

[JoseMorenoofi]

Links are cut, please, can you share the links? Thanks

[selena-mq]

This site now got Unexpected build error...