speechsr48k/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": 32,
      "fp16_run": false,
      "lr_decay": 0.995,
      "segment_size": 9600,
      "init_lr_ratio": 1,
      "warmup_epochs": 0,
      "c_mel": 45
    },
    "data": {
      "train_filelist_path": "filelists/train_48k_vctk_trim_bigvgan_sr.txt",
      "test_filelist_path": "filelists/test_48k_vctk_trim_bigvgan_sr.txt",
      "text_cleaners":["english_cleaners2"],
      "max_wav_value": 32768.0,
      "sampling_rate": 48000,
      "filter_length": 1280,
      "hop_length": 320,
      "win_length": 1280,
      "n_mel_channels": 128,
      "mel_fmin": 0,
      "mel_fmax": 24000,
      "add_blank": true,
      "n_speakers": 0,
      "cleaned_text": true,
      "aug_rate": 1.0,
      "top_db": 20
    },
    "model": {
      "resblock": "0",
      "resblock_kernel_sizes": [3,7,11],
      "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
      "upsample_rates": [3],
      "upsample_initial_channel": 32,
      "upsample_kernel_sizes": [3],
      "use_spectral_norm": false
    }
  }