Upload 13 files

speaker_diarization/base.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed3a0b6b1c0edf879ad9b11b1af5a0e6ab5db9205f891f668f8b0e6c6326e34e
+size 145262807

speaker_diarization/large-v1.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4b87e7e0bf463eb8e6956e646f1e277e901512310def2c24bf0e11bd3c28e9a
+size 3086999982

speaker_diarization/large-v2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524
+size 3086999982

speaker_diarization/large-v3-turbo.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aff26ae408abcba5fbf8813c21e62b0941638c5f6eebfb145be0c9839262a19a
+size 1617941637

speaker_diarization/large-v3.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5b1a55b89c1367dacf97e3e19bfd829a01529dbfdeefa8caeb59b3f1b81dadb
+size 3087371615

speaker_diarization/medium.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1
+size 1528008539

speaker_diarization/small.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794
+size 483617219

speaker_diarization/speechbrain/classifier.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd9e3634fe68bd0a427c95e354c0c677374f62b3f434e45b78599950d860d535
+size 5534328

speaker_diarization/speechbrain/config.json

@@ -0,0 +1,3 @@
+{
+  "speechbrain_interface": "SpeakerRecognition"
+}

speaker_diarization/speechbrain/embedding_model.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0575cb64845e6b9a10db9bcb74d5ac32b326b8dc90352671d345e2ee3d0126a2
+size 83316686

speaker_diarization/speechbrain/hyperparams.yaml

@@ -0,0 +1,58 @@
+# ############################################################################
+# Model: ECAPA big for Speaker verification
+# ############################################################################
+
+# Feature parameters
+n_mels: 80
+
+# Pretrain folder (HuggingFace)
+pretrained_path: assets/models/speaker_diarization/models/speechbrain
+
+# Output parameters
+out_n_neurons: 7205
+
+# Model params
+compute_features: !new:main.library.speaker_diarization.features.Fbank
+    n_mels: !ref <n_mels>
+
+mean_var_norm: !new:main.library.speaker_diarization.features.InputNormalization
+    norm_type: sentence
+    std_norm: False
+
+embedding_model: !new:main.library.speaker_diarization.ECAPA_TDNN.ECAPA_TDNN
+    input_size: !ref <n_mels>
+    channels: [1024, 1024, 1024, 1024, 3072]
+    kernel_sizes: [5, 3, 3, 3, 1]
+    dilations: [1, 2, 3, 4, 1]
+    attention_channels: 128
+    lin_neurons: 192
+
+classifier: !new:main.library.speaker_diarization.ECAPA_TDNN.Classifier
+    input_size: 192
+    out_neurons: !ref <out_n_neurons>
+
+mean_var_norm_emb: !new:main.library.speaker_diarization.features.InputNormalization
+    norm_type: global
+    std_norm: False
+
+modules:
+    compute_features: !ref <compute_features>
+    mean_var_norm: !ref <mean_var_norm>
+    embedding_model: !ref <embedding_model>
+    mean_var_norm_emb: !ref <mean_var_norm_emb>
+    classifier: !ref <classifier>
+        
+label_encoder: !new:main.library.speaker_diarization.encoder.CategoricalEncoder
+
+        
+pretrainer: !new:main.library.speaker_diarization.parameter_transfer.Pretrainer
+    loadables:
+        embedding_model: !ref <embedding_model>
+        mean_var_norm_emb: !ref <mean_var_norm_emb>
+        classifier: !ref <classifier>
+        label_encoder: !ref <label_encoder>
+    paths:
+        embedding_model: !ref <pretrained_path>/embedding_model.ckpt
+        mean_var_norm_emb: !ref <pretrained_path>/mean_var_norm_emb.ckpt
+        classifier: !ref <pretrained_path>/classifier.ckpt
+        label_encoder: !ref <pretrained_path>/label_encoder.txt

speaker_diarization/speechbrain/mean_var_norm_emb.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd70225b05b37be64fc5a95e24395d804231d43f74b2e1e5a513db7b69b34c33
+size 1921

speaker_diarization/tiny.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65147644a518d12f04e32d6f3b26facc3f8dd46e5390956a9424a650c0ce22b9
+size 75572083