configs/config.json

{
    "data": {
        "add_blank": true,
        "filter_length": 1024,
        "hop_length": 256,
        "max_wav_value": 32768.0,
        "mel_fmax": null,
        "mel_fmin": 0.0,
        "n_mel_channels": 80,
        "n_speakers": 6,
        "sampling_rate": 22050,
        "text_cleaners": [ "english_cleaners2" ],
        "training_files": "filelists/uma_train_filelist.txt",
        "validation_files": "filelists/uma_val_filelist.txt",
        "win_length": 1024
    },
    "model": {
        "filter_channels": 768,
        "gin_channels": 256,
        "hidden_channels": 256,
        "inter_channels": 192,
        "kernel_size": 3,
        "n_heads": 2,
        "n_layers": 6,
        "n_layers_q": 3,
        "p_dropout": 0.1,
        "resblock": "1",
        "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
        "resblock_kernel_sizes": [3,7,11],
        "upsample_initial_channel": 512,
        "upsample_kernel_sizes": [16,16,4,4],
        "upsample_rates": [8,8,2,2],
        "use_spectral_norm": false
    },
    "speakers": [ "\u4e1c\u6d77\u5e1d\u738b", "\u4e4c\u62c9\u62c9", "\u5f85\u517c\u798f\u6765", "\u5f85\u517c\u8bd7\u6b4c\u5267", "onikuru", "cpop" ],
    "train": {
        "batch_size": 40,
        "betas": [0.8, 0.99],
        "c_kl": 1.0,
        "c_mel": 45,
        "epochs": 10000,
        "eps": 1e-09,
        "eval_interval": 2000,
        "fp16_run": true,
        "init_lr_ratio": 1,
        "learning_rate": 0.0002,
        "log_interval": 200,
        "lr_decay": 0.999875,
        "seed": 1234,
        "segment_size": 8192,
        "warmup_epochs": 0
    }
}