How to deal with the speaker_turn_scores of the constraint_matrix ?

wq2012 / SpectralCluster

Python re-implementation of the (constrained) spectral clustering algorithms used in Google's speaker diarization papers.

https://google.github.io/speaker-id/publications/LstmDiarization/

Apache License 2.0

513 stars 73 forks source link

How to deal with the speaker_turn_scores of the constraint_matrix ? #34

Closed LONG520520 closed 3 years ago

LONG520520 commented 3 years ago

Hello, I'd like to ask if you can give some suggestions on how to get the speaker order score in this restriction matrix? Or whether a template can be provided.

''' from spectralcluster import constraint constraint_matrix = constraint.ConstraintMatrix( speaker_turn_scores, threshold=1).compute_diagonals() ''' Thank you.

wq2012 commented 3 years ago

See #32

Unfortunately we cannot open source our speaker turn detection algorithm due to dependencies.