diff-vits/config.json at master · adelacvg/diff-vits · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
{
  "train": {
    "train_batch_size":32,
    "gradient_accumulate_every": 1,
    "train_lr": 0.0001,
    "train_num_steps": 1000000,
    "adam_betas": [0.9, 0.99],
    "save_and_sample_every":1000,
    "timesteps":1000,
    "logs_folder" : "logs/tts",
    "seed": 1234,
    "epochs": 10000,
    "use_noise_scaled_mas": true,
    "mas_noise_scale_initial":0.01,
    "noise_scale_delta":2e-6,
    "betas": [
      0.8,
      0.99
    ],
    "num_workers": 32,
    "eps": 1e-09,
    "keep_ckpts": 3
  },
  "data": {
    "training_files": "../AISHELL3_mas_processed",
    "val_files": "dataset_processed",
    "sampling_rate": 24000,
    "hop_length": 256,
    "window_size":1024,
    "language" : "zh",
    "add_blank" : true,
    "min_text_len": 1,
    "max_text_len": 300,
    "win_length": 2048,
    "n_mel_channels": 100,
    "mel_fmin": 0.0,
    "mel_fmax": null,
    "cleaned_text": true
  },
  "diffusion_encoder":{
    "in_channels":100,
    "out_channels":100,
    "hidden_channels":128,
    "n_heads":8,
    "p_dropout":0.2
  },
  "vits":{
    "use_spk_conditioned_encoder": true,
    "use_noise_scaled_mas": true,
    "use_mel_posterior_encoder": false,
    "use_duration_discriminator": true,
    "inter_channels": 128,
    "hidden_channels": 256,
    "filter_channels": 256,
    "n_heads": 2,
    "n_layers": 6,
    "kernel_size": 3,
    "p_dropout": 0.1,
    "n_layers_q": 4,
    "use_spectral_norm": false,
    "gin_channels": 256
  }
}