| | |
| | |
| | |
| |
|
| | |
| | n_mels: 80 |
| |
|
| | |
| | pretrained_path: assets/models/speaker_diarization/models/speechbrain |
| |
|
| | |
| | out_n_neurons: 7205 |
| |
|
| | |
| | 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 |
| |
|
