Speaker diarization pipeline in pyannote 2.0 ?

[Gabrielk99]

Hi, thanks for the suport every ask that i do here.

In my recents issues you recommend me to use the segmentation model from pyannote 2.0, well, i read the readme and the hub page, and i have some questions.

My problem is the Diarization problem, i need the step of segmentation and the step of clustering and classification of speakers.

So

1. How i can use the model Segmentation from pyannote 2.0 with the Diarization Model? In pyannote 2.0 has a tutorial with that? Sorry if this question has already been made, just let me know and i research for that.

[hbredin]

pyannote 2.0 does not have a working speaker diarization pipeline yet, only pyannote 1.1 does.

[Gabrielk99]

Oh but i can use pyannote 2.0 for the task? I see that Segmentation model return different waves to recognize the moment of activity voice, every one is a different speaker?

I'm trying use the Segmentation model, and i cant print the output

this error mensage : ValueError: too many values to unpack (expected 2)

[hbredin]

The segmentation model can be used for diarization of 5s chunk but cannot be used directly to perform diarization for a whole file.
Please read the related paper for a better understand of what this does.

[Gabrielk99]

Thanks, i will read the paper!!