logs/hierspeechpp_eng_kor/config.json

{
    "train": {
      "log_interval": 200,
      "eval_interval": 10000,
      "save_interval": 10000,
      "seed": 1234,
      "epochs": 20000,
      "learning_rate": 1e-4,
      "betas": [0.8, 0.99],
      "eps": 1e-9,
      "batch_size": 20,
      "fp16_run": false,
      "lr_decay": 0.999,
      "segment_size": 61440,
      "init_lr_ratio": 1,
      "warmup_epochs": 0,
      "c_mel": 45,
      "c_kl": 1.0,
      "c_bi_kl": 0.5,
      "c_mixup": 1,
      "c_pho": 45.0,
      "c_f0": 45,
      "aug": true
    },
    "data": {
      "train_filelist_path": "filelists/train_eng_kor_wav.txt",
      "test_filelist_path": "filelists/test_hier_16k_72_wav.txt",
      "text_cleaners":["english_cleaners2"],
      "max_wav_value": 32768.0,
      "sampling_rate": 16000,
      "filter_length": 1280,
      "hop_length": 320,
      "win_length": 1280,
      "n_mel_channels": 80,
      "mel_fmin": 0,
      "mel_fmax": 8000,
      "add_blank": true,
      "n_speakers": 0,
      "cleaned_text": true,
      "aug_rate": 1.0,
      "top_db": 20
    },
    "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": [4,5,4,2,2],
      "upsample_initial_channel": 512,
      "upsample_kernel_sizes": [8,11,8,4,4],
      "n_layers_q": 3,
      "use_spectral_norm": false,
      "wav2vec_feature_layer": 7,
      "gin_channels": 256,
      "cfg": true,
      "uncond_ratio": 0.1,
      "prosody_size": 20
  
    }
  }