{ "train": { "log_interval": 200, "eval_interval": 1000, "seed": 1234, "epochs": 10000, "learning_rate": 0.0001, "betas": [ 0.8, 0.99 ], "eps": 1e-09, "batch_size": 32, "fp16_run": true, "lr_decay": 0.999875, "segment_size": 8192, "init_lr_ratio": 1, "warmup_epochs": 0, "c_mel": 45, "c_kl": 1.0 }, "data": { "training_files": "filelists/train.list", "validation_files": "filelists/val.list", "max_wav_value": 32768.0, "sampling_rate": 22050, "filter_length": 1024, "hop_length": 256, "win_length": 1024, "n_mel_channels": 80, "mel_fmin": 0.0, "mel_fmax": null, "add_blank": true, "n_speakers": 300, "cleaned_text": true, "spk2id": { "SSB0080": 0, "SSB0012": 1, "SSB0038": 2, "SSB0382": 3, "SSB0394": 4, "SSB0395": 5, "SSB0316": 6, "SSB0200": 7, "SSB1408": 8, "SSB1392": 9, "SSB0700": 10, "SSB1138": 11, "SSB1072": 12, "SSB0751": 13, "SSB0338": 14, "SSB0435": 15, "SSB0913": 16, "SSB1806": 17, "SSB1878": 18, "SSB1385": 19, "SSB0817": 20, "SSB0599": 21, "SSB0887": 22, "SSB0720": 23, "SSB1091": 24, "SSB0786": 25, "SSB0737": 26, "SSB0666": 27, "SSB0606": 28, "SSB0535": 29, "SSB0112": 30, "SSB1161": 31, "SSB1448": 32, "SSB1684": 33, "SSB1699": 34, "SSB1341": 35, "SSB0919": 36, "SSB1056": 37, "SSB1115": 38, "SSB1563": 39, "SSB0482": 40, "SSB0502": 41, "SSB0415": 42, "SSB0145": 43, "0001_Angry": 44, "0001_Happy": 45, "0001_Neutral": 46, "0001_Sad": 47, "0001_Surprise": 48, "0002_Angry": 49, "0002_Happy": 50, "0002_Neutral": 51, "0002_Sad": 52, "0002_Surprise": 53, "0003_Angry": 54, "0003_Happy": 55, "0003_Neutral": 56, "0003_Sad": 57, "0003_Surprise": 58, "0004_Angry": 59, "0004_Happy": 60, "0004_Neutral": 61, "0004_Sad": 62, "0004_Surprise": 63, "0005_Angry": 64, "0005_Happy": 65, "0005_Neutral": 66, "0005_Sad": 67, "0005_Surprise": 68, "0006_Angry": 69, "0006_Happy": 70, "0006_Neutral": 71, "0006_Sad": 72, "0006_Surprise": 73, "0007_Angry": 74, "0007_Happy": 75, "0007_Neutral": 76, "0007_Sad": 77, "0007_Surprise": 78, "0008_Angry": 79, "0008_Happy": 80, "0008_Neutral": 81, "0008_Sad": 82, "0008_Surprise": 83, "0009_Angry": 84, "0009_Happy": 85, "0009_Neutral": 86, "0009_Sad": 87, "0009_Surprise": 88, "0010_Angry": 89, "0010_Happy": 90, "0010_Neutral": 91, "0010_Sad": 92, "0010_Surprise": 93, "0011_Angry": 94, "0011_Happy": 95, "0011_Neutral": 96, "0011_Sad": 97, "0011_Surprise": 98, "0012_Angry": 99, "0012_Happy": 100, "0012_Neutral": 101, "0012_Sad": 102, "0012_Surprise": 103, "0013_Angry": 104, "0013_Happy": 105, "0013_Neutral": 106, "0013_Sad": 107, "0013_Surprise": 108, "0014_Angry": 109, "0014_Happy": 110, "0014_Neutral": 111, "0014_Sad": 112, "0014_Surprise": 113, "0015_Angry": 114, "0015_Happy": 115, "0015_Neutral": 116, "0015_Sad": 117, "0015_Surprise": 118, "0016_Angry": 119, "0016_Happy": 120, "0016_Neutral": 121, "0016_Sad": 122, "0016_Surprise": 123, "0017_Angry": 124, "0017_Happy": 125, "0017_Neutral": 126, "0017_Sad": 127, "0017_Surprise": 128, "0018_Angry": 129, "0018_Happy": 130, "0018_Neutral": 131, "0018_Sad": 132, "0018_Surprise": 133, "0019_Angry": 134, "0019_Happy": 135, "0019_Neutral": 136, "0019_Sad": 137, "0019_Surprise": 138, "0020_Angry": 139, "0020_Happy": 140, "0020_Neutral": 141, "0020_Sad": 142, "0020_Surprise": 143 } }, "model": { "inter_channels": 192, "hidden_channels": 192, "filter_channels": 768, "n_heads": 2, "n_layers": 6, "kernel_size": 3, "p_dropout": 0.1, "resblock": "1", "resblock_kernel_sizes": [ 3, 7, 11 ], "resblock_dilation_sizes": [ [ 1, 3, 5 ], [ 1, 3, 5 ], [ 1, 3, 5 ] ], "upsample_rates": [ 8, 8, 2, 2 ], "upsample_initial_channel": 512, "upsample_kernel_sizes": [ 16, 16, 4, 4 ], "n_layers_q": 3, "use_spectral_norm": false, "gin_channels": 256 } }