File size: 3,599 Bytes
d1b91e7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f26c987
d1b91e7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
accumulate_grad_batches: 1
amp: false
audio_num_mel_bins: 80
audio_sample_rate: 22050
base_config:
- egs/egs_bases/tts/fs2_orig.yaml
- ./base_text2mel.yaml
binarization_args:
  min_sil_duration: 0.1
  shuffle: false
  test_range:
  - 0
  - 523
  train_range:
  - 871
  - -1
  trim_eos_bos: false
  valid_range:
  - 523
  - 871
  with_align: true
  with_f0: true
  with_f0cwt: true
  with_linear: false
  with_spk_embed: false
  with_wav: false
binarizer_cls: data_gen.tts.base_binarizer.BaseBinarizer
binary_data_dir: data/binary/ljspeech_cwt
check_val_every_n_epoch: 10
clip_grad_norm: 1
clip_grad_value: 0
conv_use_pos: false
cwt_std_scale: 1.0
debug: false
dec_dilations:
- 1
- 1
- 1
- 1
dec_ffn_kernel_size: 9
dec_inp_add_noise: false
dec_kernel_size: 5
dec_layers: 4
dec_post_net_kernel: 3
decoder_rnn_dim: 0
decoder_type: fft
dropout: 0.0
ds_workers: 2
dur_predictor_kernel: 3
dur_predictor_layers: 2
enc_dec_norm: ln
enc_dilations:
- 1
- 1
- 1
- 1
enc_ffn_kernel_size: 9
enc_kernel_size: 5
enc_layers: 4
enc_post_net_kernel: 3
enc_pre_ln: true
enc_prenet: true
encoder_K: 8
encoder_type: fft
endless_ds: true
eval_max_batches: -1
f0_max: 600
f0_min: 80
ffn_act: gelu
ffn_hidden_size: 1024
fft_size: 1024
fmax: 7600
fmin: 80
frames_multiple: 1
gen_dir_name: ''
griffin_lim_iters: 30
hidden_size: 256
hop_size: 256
infer: false
lambda_commit: 0.25
lambda_energy: 0.1
lambda_f0: 1.0
lambda_ph_dur: 0.1
lambda_sent_dur: 1.0
lambda_uv: 1.0
lambda_word_dur: 1.0
layers_in_block: 2
load_ckpt: ''
loud_norm: false
lr: 0.0005
max_epochs: 1000
max_frames: 1548
max_input_tokens: 1550
max_sentences: 128
max_tokens: 40000
max_updates: 160000
max_valid_sentences: 1
max_valid_tokens: 60000
mel_losses: l1:0.5|ssim:0.5
mel_vmax: 1.5
mel_vmin: -6
min_frames: 0
num_ckpt_keep: 3
num_heads: 2
num_sanity_val_steps: 5
num_spk: 1
num_valid_plots: 10
optimizer_adam_beta1: 0.9
optimizer_adam_beta2: 0.98
out_wav_norm: false
pitch_extractor: parselmouth
pitch_key: pitch
pitch_type: cwt
predictor_dropout: 0.5
predictor_grad: 0.1
predictor_hidden: -1
predictor_kernel: 5
predictor_layers: 2
preprocess_args:
  add_eos_bos: true
  mfa_group_shuffle: false
  mfa_offset: 0.02
  nsample_per_mfa_group: 1000
  reset_phone_dict: true
  reset_word_dict: true
  save_sil_mask: true
  txt_processor: en
  use_mfa: true
  vad_max_silence_length: 12
  wav_processors: []
  with_phsep: true
preprocess_cls: egs.datasets.audio.lj.preprocess.LJPreprocess
print_nan_grads: false
processed_data_dir: data/processed/ljspeech
profile_infer: false
raw_data_dir: data/raw/LJSpeech-1.1
ref_norm_layer: bn
rename_tmux: true
resume_from_checkpoint: 0
save_best: false
save_codes:
- tasks
- modules
- egs
save_f0: false
save_gt: true
scheduler: warmup
seed: 1234
sort_by_len: true
task_cls: tasks.tts.fs2_orig.FastSpeech2OrigTask
tb_log_interval: 100
test_ids:
- 0
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 68
- 70
- 74
- 87
- 110
- 172
- 190
- 215
- 231
- 294
- 316
- 324
- 402
- 422
- 485
- 500
- 505
- 508
- 509
- 519
test_input_yaml: ''
test_num: 100
test_set_name: test
train_set_name: train
train_sets: ''
use_energy_embed: true
use_gt_dur: false
use_gt_energy: false
use_gt_f0: false
use_pitch_embed: true
use_pos_embed: true
use_spk_embed: false
use_spk_id: false
use_uv: true
use_word_input: false
val_check_interval: 2000
valid_infer_interval: 10000
valid_monitor_key: val_loss
valid_monitor_mode: min
valid_set_name: valid
vocoder: HifiGAN
vocoder_ckpt: checkpoints/hifi_lj
warmup_updates: 4000
weight_decay: 0
win_size: 1024
word_dict_size: 10000
work_dir: checkpoints/fs2_exp