fastspeech_squeezewave/SqueezeWave/configs/config_a128_c128.json

{
    "train_config": {
        "fp16_run": true,
        "output_directory": "checkpoints",
        "epochs": 100000,
        "learning_rate": 4e-4,
        "sigma": 1.0,
        "iters_per_checkpoint": 2000,
        "batch_size": 96, 
        "seed": 1234,
        "checkpoint_path": "",
        "with_tensorboard": true
    },
    "data_config": {
        "training_files": "train_files.txt",
        "segment_length": 16384,
        "sampling_rate": 22050,
        "filter_length": 1024,
        "hop_length": 256,
        "win_length": 1024,
        "mel_fmin": 0.0,
        "mel_fmax": 8000.0
    },
    "dist_config": {
        "dist_backend": "nccl",
        "dist_url": "tcp://localhost:54321"
    },
    "squeezewave_config": {
        "n_mel_channels": 80,
        "n_flows": 12,
        "n_audio_channel": 128,
        "n_early_every": 2,
        "n_early_size": 16,
        "WN_config": {
            "n_layers": 8,
            "n_channels": 128,
            "kernel_size": 3
        }
    }
}