File size: 1,331 Bytes
a6fd2ee | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 | {
"_class_name": "MotifAudio",
"auto_map": {
"AutoModel": "model.MotifAudio"
},
"sample_rate": 16000,
"semantic_config": {
"norm_type": "global",
"norm_mean": -7.115164,
"norm_std": 4.229471,
"sem_dim": 1024,
"n_mels": 80,
"n_fft": 400,
"hop_length": 160,
"f_min": 50.0,
"f_max": 8000.0,
"patch_f": 16,
"patch_t": 16,
"semantic_encoder": {
"img_size": [
80,
512
],
"patch_size": [
16,
16
],
"in_chans": 1,
"embed_dim": 1024,
"depth": 48,
"num_heads": 16,
"mlp_ratio": 4.0,
"use_cls_token": false,
"num_register_tokens": 4,
"use_rope": true,
"rope_theta": 100.0,
"layer_scale_init": 0.0
}
},
"acoustic_config": {
"sample_rate": 16000,
"n_fft": 1024,
"hop_length": 160,
"n_mels": 160,
"patch_f": 160,
"patch_t": 2,
"embed_dim": 1024,
"f_min": 50.0,
"f_max": 8000.0,
"norm_mean": -7.115164,
"norm_std": 4.229471
},
"fusion_config": {
"dim": 1024,
"num_heads": 8
},
"decoder_config": {
"input_channels": 1024,
"hidden_dim": 1024,
"intermediate_dim": 4096,
"num_layers": 12,
"n_fft": 1024,
"hop_length": 160,
"sample_rate": 16000,
"upsample_tokens": 2
}
}
|