kamilakesbi commited on
Commit
d6d2a6e
1 Parent(s): 20e77f0

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. README.md +1 -1
  2. config.yaml +18 -7
  3. pytorch_model.bin +2 -2
README.md CHANGED
@@ -20,6 +20,6 @@ tags:
20
  - speaker-embedding
21
  - PyTorch
22
  - wespeaker
23
- licence: cc-by-4.0
24
  ---
25
  This is the model card of a pyannote model that has been pushed on the Hub. This model card has been automatically generated.
 
20
  - speaker-embedding
21
  - PyTorch
22
  - wespeaker
23
+ licence: mit
24
  ---
25
  This is the model card of a pyannote model that has been pushed on the Hub. This model card has been automatically generated.
config.yaml CHANGED
@@ -1,10 +1,21 @@
1
  model:
2
- _target_: pyannote.audio.models.embedding.wespeaker.WeSpeakerResNet34
3
- dither: 0.0
4
- frame_length: 25
5
- frame_shift: 10
 
 
 
 
 
 
 
6
  num_channels: 1
7
- num_mel_bins: 80
8
  sample_rate: 16000
9
- use_energy: false
10
- window_type: hamming
 
 
 
 
 
 
1
  model:
2
+ _target_: pyannote.audio.models.segmentation.PyanNet.PyanNet
3
+ linear:
4
+ hidden_size: 128
5
+ num_layers: 2
6
+ lstm:
7
+ batch_first: true
8
+ bidirectional: true
9
+ dropout: 0.0
10
+ hidden_size: 128
11
+ monolithic: true
12
+ num_layers: 4
13
  num_channels: 1
 
14
  sample_rate: 16000
15
+ sincnet:
16
+ sample_rate: 16000
17
+ stride: 10
18
+ task:
19
+ duration: 10.0
20
+ max_speakers_per_chunk: 3
21
+ max_speakers_per_frame: 2
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c6ebcd5526d688deb24b12eaae4774a6d8337c22bf7b7f25c1eaa1c345f6cae5
3
- size 26644594
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf667e302cb3ad72316803868e2cf007d35d506e4ac6daafdd527dfd69f3fa72
3
+ size 5912144