-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathconfig.yaml
127 lines (110 loc) · 2.79 KB
/
config.yaml
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
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
exp_name: 'multi'
gpu: 'cpu'
run_debug_eval: false
logger:
offline: false
wandb_key:
tts:
weights_path: './pretrained/290000.pth.tar'
restore_step: 0
hifi:
weights_path: './pretrained/hifi.pth'
MAX_WAV_VALUE: 32768
resblock: "1"
num_gpus: 0
batch_size: 8
learning_rate: 0.0002
adam_b1: 0.8
adam_b2: 0.99
lr_decay: 0.999
seed: 1234
upsample_rates: [8,8,2,2]
upsample_kernel_sizes: [16,16,4,4]
upsample_initial_channel: 512
resblock_kernel_sizes: [3,7,11]
resblock_dilation_sizes: [[1,3,5], [1,3,5], [1,3,5]]
resblock_initial_channel: 256
segment_size: 8192
num_mels: 80
num_freq: 1025
n_fft: 1024
hop_size: 256
win_size: 1024
sampling_rate: 22050
train_config:
path:
ckpt_path: "../output/ckpt/multi_final"
result_path: "../output/result/multi_final"
optimizer:
batch_size: 16
betas: [0.95, 0.999]
eps: 0.00001
weight_decay: 0.0
grad_clip_thresh: 1.0
grad_acc_step: 4
warm_up_step: 4000
anneal_steps: [300000, 400000, 500000]
anneal_rate: 0.7
step:
total_step: 900000
log_step: 100
synth_step: 1000
val_step: 1000
save_step: 5000
max_masks_per_sentence: 0.15
preprocess_config:
dataset: "MAIN"
path:
lexicon_path: "./rus_all.dict"
raw_path: "./speakers"
preprocessed_path: "./processed"
preprocessing:
val_size: 512
text:
text_cleaners: []
language: "ru"
audio:
sampling_rate: 22050
max_wav_value: 32768.0
stft:
filter_length: 1024
hop_length: 256
win_length: 1024
mel:
n_mel_channels: 80
mel_fmin: 0
mel_fmax: 8000 # please set to 8000 for HiFi-GAN vocoder, set to null for MelGAN vocoder
pitch:
feature: "phoneme_level" # support 'phoneme_level' or 'frame_level'
normalization: True
energy:
feature: "phoneme_level" # support 'phoneme_level' or 'frame_level'
normalization: True
model_config:
transformer:
encoder_layer: 4
encoder_head: 2
encoder_hidden: 256
variance_hidden: 256
decoder_layer: 6
decoder_head: 2
decoder_hidden: 256
conv_filter_size: 1024
conv_kernel_size: [9, 1]
encoder_dropout: 0.2
decoder_dropout: 0.2
variance_predictor:
filter_size: 256
kernel_size: 3
dropout: 0.5
use_cwt: False
variance_embedding:
pitch_quantization: "linear" # support 'linear' or 'log', 'log' is allowed only if the pitch values are not normalized during preprocessing
energy_quantization: "linear" # support 'linear' or 'log', 'log' is allowed only if the energy values are not normalized during preprocessing
n_bins: 256
multi_speaker: True
max_seq_len: 1000
vocoder:
model: "HiFi-GAN" # support 'HiFi-GAN', 'MelGAN'
speaker: "universal" # support 'LJSpeech', 'universal'
use_cpu: true