config.json

{
  "model_params": {
    "decoder": {
      "resblock_dilation_sizes": [
        [
          1,
          3,
          5
        ],
        [
          1,
          3,
          5
        ],
        [
          1,
          3,
          5
        ]
      ],
      "resblock_kernel_sizes": [
        3,
        7,
        11
      ],
      "type": "hifigan",
      "upsample_initial_channel": 512,
      "upsample_kernel_sizes": [
        20,
        10,
        6,
        4
      ],
      "upsample_rates": [
        10,
        5,
        3,
        2
      ]
    },
    "diffusion": {
      "dist": {
        "estimate_sigma_data": true,
        "mean": -3.0,
        "sigma_data": 0.2,
        "std": 1.0
      },
      "embedding_mask_proba": 0.1,
      "transformer": {
        "head_features": 64,
        "multiplier": 2,
        "num_heads": 8,
        "num_layers": 3
      }
    },
    "dim_in": 64,
    "dropout": 0.2,
    "hidden_dim": 512,
    "max_conv_dim": 512,
    "max_dur": 50,
    "multispeaker": true,
    "n_layer": 3,
    "n_mels": 80,
    "n_token": 178,
    "slm": {
      "hidden": 768,
      "initial_channel": 64,
      "model": "microsoft/wavlm-base-plus",
      "nlayers": 13,
      "sr": 16000
    },
    "style_dim": 128
  },
  "training_config": {
    "epochs": 5,
    "batch_size": 2,
    "max_len": 600,
    "optimizer": {
      "bert_lr": 1e-05,
      "ft_lr": 0.0001,
      "lr": 0.0001
    },
    "loss_params": {
      "diff_epoch": 2,
      "joint_epoch": 110,
      "lambda_F0": 1.0,
      "lambda_ce": 20.0,
      "lambda_diff": 1.0,
      "lambda_dur": 1.0,
      "lambda_gen": 1.0,
      "lambda_mel": 5.0,
      "lambda_mono": 1.0,
      "lambda_norm": 1.0,
      "lambda_s2s": 1.0,
      "lambda_slm": 1.0,
      "lambda_sty": 1.0
    }
  },
  "preprocess_params": {
    "spect_params": {
      "hop_length": 300,
      "n_fft": 2048,
      "win_length": 1200
    },
    "sr": 24000
  },
  "data_params": {
    "OOD_data": "Data/OOD_texts.txt",
    "min_length": 50,
    "root_path": "Data/wavs",
    "train_data": "Data/train_list.txt",
    "val_data": "Data/val_list.txt"
  },
  "model_state": {
    "epoch": 3,
    "iterations": 3384,
    "val_loss": 0.39982396364212036
  },
  "training_metrics": {
    "train_loss": [],
    "val_loss": [
      18.0,
      30.0,
      30.0,
      44.0,
      33.0,
      21.0,
      17.0,
      44.0,
      45.0,
      55.0,
      29.0,
      8.0,
      45.0,
      29.0,
      11.0,
      52.0,
      48.0,
      18.0,
      5.0,
      4.0,
      57.0,
      9.0
    ],
    "dur_loss": [
      0.44,
      0.398,
      0.368,
      0.373,
      0.362,
      0.371,
      0.376,
      0.438,
      0.425,
      0.418,
      0.41,
      0.4,
      0.401,
      0.399,
      0.409,
      0.397,
      0.408,
      0.432,
      0.404,
      0.406,
      0.4,
      0.403
    ],
    "F0_loss": [
      2.865,
      2.754,
      2.769,
      2.916,
      2.745,
      2.811,
      2.793,
      2.665,
      2.761,
      2.683,
      2.783,
      2.524,
      2.615,
      2.515,
      2.663,
      2.66,
      2.477,
      1.751,
      1.739,
      1.76,
      1.691,
      1.72
    ],
    "epochs": [
      1,
      2,
      3,
      4,
      5,
      6,
      7,
      8,
      9,
      10,
      11,
      12,
      13,
      14,
      15,
      16,
      17,
      18,
      19,
      20,
      21,
      22
    ]
  }
}