continuous_64ch.json

{
    "name": "DiffWave",
    "arch": {
        "type": "DiffWave",
        "args": {
            "T": 1000
        }
    },
    "dataset": {
        "type": "RandomWAVDataset",
        "args": {
            "data_dir": "~/data-disk/Datasets/LJ/LJSpeech-1.1/wavs/",
            "size": 12000000,
            "segment": 16000,
            "deterministic": false
        }
    },
    "optimizer": {
        "type": "Adam",
        "args": {
            "lr": 2e-4
        }
    },
    "lr_scheduler": {
        "type": "ReduceLROnPlateau",
        "args": {
            "factor": 0.3,
            "patience": 1000000,
            "verbose": true
        }
    },
    "data_loader": {
        "batch_size": 12,
        "shuffle": false,
        "drop_last": false,
        "num_workers": 4,
        "prefetch_factor": 4,
        "pin_memory": true
    },
    "trainer": {
        "save_dir": "saved/",
        "log_dir": "runs/",
        "eval_file": "~/data-disk/Datasets/LJ/LJSpeech-1.1/wavs/LJ001-0001.wav",
        "cum_steps": 1,
        "n_fft": 1024,
        "hop_length": 256,
        "n_mels": 80,
        "sr": 22050,
        "eval_interval": 10000,
        "train_T": 0,
        "eval_T": 50,
        "extra_monitor": [
            "kld",
            "ll",
            "loss_T"
        ],
        "with_amp": true,
        "max_log_snr": 23,
        "min_log_snr": -3.6
    }
}