ValueError: Found array with dim 3. AgglomerativeClustering expected <= 2. about simplediarization

[mirix]

Thanks, I believe the problem is using Python 3.11.

In any case, I have come up with an entirely different procedure:

https://github.com/mirix/approaches-to-diarisation

[yejinc00]

I think this issue might be due to the number of channels in the audio files.
In my case, I resolved it by using ffmpeg to adjust the channel count to one.

ffmpeg -i input_stereo.wav -ac 1 output_mono.wav
(Mono=1, Stereo=2)

[JaesungHuh]

Thank you for pointing this out. I changed the README based on your comment.

[JaesungHuh]

Hi, I ran the diarization with the youtube wavfile and it shows me the corresponding results. This is the result with num_cluster = 3 and I checked num_cluster = None also works. (although it gives shit results ahaha)

Could you share the config file that you used?

SPEAKER sample_out 1 0.008 0.102 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 1.679 11.000 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 12.679 1.500 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 14.179 9.000 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 23.179 1.855 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 26.249 2.500 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 28.749 4.824 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 34.130 6.564 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 40.863 2.396 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 44.947 0.978 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 47.208 0.607 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 48.693 2.000 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 50.693 1.000 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 51.693 4.627 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 57.029 1.000 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 58.029 3.928 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 62.598 2.801 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 65.669 0.219 <NA> <NA> 1 <NA> <NA>
SPEAKER sample_out 1 66.952 2.058 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 69.432 25.330 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 96.415 1.620 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 98.693 11.000 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 109.693 3.500 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 113.193 2.500 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 115.693 3.500 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 119.193 1.792 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 121.120 2.076 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 123.972 5.063 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 131.279 0.793 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 132.292 3.000 <NA> <NA> 0 <NA> <NA>
SPEAKER sample_out 1 135.292 6.399 <NA> <NA> 2 <NA> <NA>
SPEAKER sample_out 1 142.940 2.565 <NA> <NA> 0 <NA> <NA>