Hugging Face's logo

hicustomer
/
pyannote-speaker-diarization

Automatic Speech Recognition
pyannote.audio
pyannote
pyannote-audio-pipeline
audio
voice
speech
speaker
speaker-diarization
speaker-change-detection
voice-activity-detection
overlapped-speech-detection
Model card Files
xet
 Community
Hervé BREDIN commited on
Commit
369ac18
·
1 Parent(s): 89e7168

feat: prepare for pyannote.audio 2.1

Browse files
Files changed (1) hide show
  1. config.yaml +10 -12
config.yaml CHANGED
@@ -1,20 +1,18 @@
1
  pipeline:
2
  name: pyannote.audio.pipelines.SpeakerDiarization
3
  params:
4
- segmentation: pyannote/segmentation@2022.07
5
- segmentation_step: 0.1
6
- segmentation_batch_size: 32
7
- # embedding: speechbrain/spkrec-ecapa-voxceleb@5c0be3875fda05e81f3c004ed8c7c06be308de1e
8
  embedding: speechbrain/spkrec-ecapa-voxceleb
9
- embedding_exclude_overlap: True
10
  embedding_batch_size: 32
11
- clustering: HiddenMarkovModelClustering
 
 
12
 
13
  params:
14
- segmentation_onset: 0.58
15
  clustering:
16
- covariance_type: diag
17
- threshold: 0.35
18
- single_cluster_detection:
19
- quantile: 0.05
20
- threshold: 1.15
 
 
1
  pipeline:
2
  name: pyannote.audio.pipelines.SpeakerDiarization
3
  params:
4
+ clustering: AgglomerativeClustering
 
 
 
5
  embedding: speechbrain/spkrec-ecapa-voxceleb
 
6
  embedding_batch_size: 32
7
+ embedding_exclude_overlap: true
8
+ segmentation: pyannote/segmentation@2022.07
9
+ segmentation_batch_size: 32
10
 
11
  params:
 
12
  clustering:
13
+ method: centroid
14
+ min_cluster_size: 15
15
+ threshold: 0.7153814381597874
16
+ segmentation:
17
+ min_duration_off: 0.5817029604921046
18
+ threshold: 0.4442333667381752