Diarization options

Parameter	Default value	Notes
cluster_type	optics	Clustering algorithm in scikit-learn to use, one of optics, dbscan, affinity, agglomerative, spectral, ``kmeans
expected_num_speakers	0	Number of speaker clusters to find, must be > 1 for agglomerative, spectral, and kmeans
sparse_threshold	0.5	Threshold on distance to limit precomputed sparse matrix

Default diarization config file

cluster_type: optics
energy_mean_scale: 0.5
max_segment_length: 30
min_pause_duration: 0.05