Upload folder using huggingface_hub

config.yaml

@@ -5,6 +5,7 @@ params:
     threshold: 0.7045654963945799
   segmentation:
     min_duration_off: 0.0
+  version: 3.1
 pipeline:
   name: pyannote.audio.pipelines.speaker_diarization.SpeakerDiarization
   params:

embedding/config.yaml

@@ -0,0 +1,10 @@
+model:
+  _target_: pyannote.audio.models.embedding.wespeaker.WeSpeakerResNet34
+  dither: 0.0
+  frame_length: 25
+  frame_shift: 10
+  num_channels: 1
+  num_mel_bins: 80
+  sample_rate: 16000
+  use_energy: false
+  window_type: hamming

embedding/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6ebcd5526d688deb24b12eaae4774a6d8337c22bf7b7f25c1eaa1c345f6cae5
+size 26644594

segmentation/config.yaml

@@ -0,0 +1,21 @@
+model:
+  _target_: pyannote.audio.models.segmentation.PyanNet.PyanNet
+  linear:
+    hidden_size: 128
+    num_layers: 2
+  lstm:
+    batch_first: true
+    bidirectional: true
+    dropout: 0.5
+    hidden_size: 128
+    monolithic: true
+    num_layers: 4
+  num_channels: 1
+  sample_rate: 16000
+  sincnet:
+    sample_rate: 16000
+    stride: 10
+task:
+  duration: 10.0
+  max_speakers_per_chunk: 3
+  max_speakers_per_frame: 2

segmentation/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20f6de79768928db1564987c6fa644ad451eb9c1139999e68336721e49f9c52c
+size 5912080