File size: 6,451 Bytes
810cc75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e40b4f0
 
 
 
 
810cc75
 
 
 
 
 
 
 
 
 
 
 
 
 
e40b4f0
810cc75
 
 
 
 
 
 
 
 
 
 
 
 
e40b4f0
 
 
810cc75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e40b4f0
810cc75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e40b4f0
810cc75
 
 
 
 
e40b4f0
810cc75
 
 
2f4c562
bc2a85a
2f4c562
 
 
 
 
 
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
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
# Generated 2022-10-08 from:
# /home/xportes/projects/speechbrain/recipes/LJSpeech/TTS/tacotron2/hparams/train.yaml
# yamllint disable
############################################################################
# Model: Tacotron2
# Tokens: Raw characters (English text)
# losses: Transducer
# Training: LJSpeech
# Authors: Georges Abous-Rjeili, Artem Ploujnikov, Yingzhi Wang
# ############################################################################


###################################
# Experiment Parameters and setup #
###################################
seed: 9234
__set_seed: !apply:torch.manual_seed [9234]
output_folder: ./results/tacotron2/9234
save_folder: ./results/tacotron2/9234/save
train_log: ./results/tacotron2/9234/train_log.txt
epochs: 750
keep_checkpoint_interval: 50

###################################
# Progress Samples                #
###################################
# Progress samples are used to monitor the progress
# of an ongoing training session by outputting samples
# of spectrograms, alignments, etc at regular intervals

# Whether to enable progress samples
progress_samples: true

# The path where the samples will be stored
progress_sample_path: ./results/tacotron2/9234/samples
# The interval, in epochs. For instance, if it is set to 5,
# progress samples will be output every 5 epochs
progress_samples_interval: 1
# The sample size for raw batch samples saved in batch.pth
# (useful mostly for model debugging)
progress_batch_sample_size: 3

#################################
# Data files and pre-processing #
#################################
data_folder: ../../../../LJSpeech-1.1
                          # e.g, /localscratch/ljspeech

train_json: ./results/tacotron2/9234/save/train.json
valid_json: ./results/tacotron2/9234/save/valid.json
test_json: ./results/tacotron2/9234/save/test.json

splits: [train, valid]
split_ratio: [90, 10]

skip_prep: false

# Use the original preprocessing from nvidia
# The cleaners to be used (applicable to nvidia only)
text_cleaners: [english_cleaners]

################################
# Audio Parameters             #
################################
sample_rate: 22050
hop_length: 256
win_length: 1024
n_mel_channels: 80
n_fft: 1024
mel_fmin: 0.0
mel_fmax: 8000.0
mel_normalized:
power: 1
norm: slaney
mel_scale: slaney
dynamic_range_compression: true

################################
# Optimization Hyperparameters #
################################
learning_rate: 0.001
weight_decay: 0.000006
batch_size: 64 #minimum 2
mask_padding: true
guided_attention_sigma: 0.2
guided_attention_weight: 50.0
guided_attention_weight_half_life: 10.
guided_attention_hard_stop: 50
gate_loss_weight: 1.0

train_dataloader_opts:
  batch_size: 64
  drop_last: false  #True #False
  num_workers: 8
  collate_fn: !new:speechbrain.lobes.models.Tacotron2.TextMelCollate

valid_dataloader_opts:
  batch_size: 64
  num_workers: 8
  collate_fn: !new:speechbrain.lobes.models.Tacotron2.TextMelCollate

test_dataloader_opts:
  batch_size: 64
  num_workers: 8
  collate_fn: !new:speechbrain.lobes.models.Tacotron2.TextMelCollate

################################
# Model Parameters and model   #
################################
n_symbols: 150 #fixed depending on symbols in textToSequence
symbols_embedding_dim: 512

# Encoder parameters
encoder_kernel_size: 5
encoder_n_convolutions: 3
encoder_embedding_dim: 512

# Decoder parameters
# The number of frames in the target per encoder step
n_frames_per_step: 1
decoder_rnn_dim: 1024
prenet_dim: 256
max_decoder_steps: 1000
gate_threshold: 0.5
p_attention_dropout: 0.1
p_decoder_dropout: 0.1
decoder_no_early_stopping: false

# Attention parameters
attention_rnn_dim: 1024
attention_dim: 128

# Location Layer parameters
attention_location_n_filters: 32
attention_location_kernel_size: 31

# Mel-post processing network parameters
postnet_embedding_dim: 512
postnet_kernel_size: 5
postnet_n_convolutions: 5

mel_spectogram: !name:speechbrain.lobes.models.Tacotron2.mel_spectogram
  sample_rate: 22050
  hop_length: 256
  win_length: 1024
  n_fft: 1024
  n_mels: 80
  f_min: 0.0
  f_max: 8000.0
  power: 1
  normalized:
  norm: slaney
  mel_scale: slaney
  compression: true

#model
model: &id002 !new:speechbrain.lobes.models.Tacotron2.Tacotron2

#optimizer
  mask_padding: true
  n_mel_channels: 80
  # symbols
  n_symbols: 150
  symbols_embedding_dim: 512
  # encoder
  encoder_kernel_size: 5
  encoder_n_convolutions: 3
  encoder_embedding_dim: 512
  # attention
  attention_rnn_dim: 1024
  attention_dim: 128
  # attention location
  attention_location_n_filters: 32
  attention_location_kernel_size: 31
  # decoder
  n_frames_per_step: 1
  decoder_rnn_dim: 1024
  prenet_dim: 256
  max_decoder_steps: 1000
  gate_threshold: 0.5
  p_attention_dropout: 0.1
  p_decoder_dropout: 0.1
  # postnet
  postnet_embedding_dim: 512
  postnet_kernel_size: 5
  postnet_n_convolutions: 5
  decoder_no_early_stopping: false

guided_attention_scheduler: &id001 !new:speechbrain.nnet.schedulers.StepScheduler
  initial_value: 50.0
  half_life: 10.

criterion: !new:speechbrain.lobes.models.Tacotron2.Loss
  gate_loss_weight: 1.0
  guided_attention_weight: 50.0
  guided_attention_sigma: 0.2
  guided_attention_scheduler: *id001
  guided_attention_hard_stop: 50

modules:
  model: *id002
opt_class: !name:torch.optim.Adam
  lr: 0.001
  weight_decay: 0.000006

#epoch object
epoch_counter: &id003 !new:speechbrain.utils.epoch_loop.EpochCounter
  limit: 750

train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
  save_file: ./results/tacotron2/9234/train_log.txt

#annealing_function
lr_annealing: &id004 !new:speechbrain.nnet.schedulers.IntervalScheduler

#infer: !name:speechbrain.lobes.models.Tacotron2.infer

  intervals:
  - steps: 6000
    lr: 0.0005
  - steps: 8000
    lr: 0.0003
  - steps: 10000
    lr: 0.0001

#checkpointer
checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
  checkpoints_dir: ./results/tacotron2/9234/save
  recoverables:
    model: *id002
    counter: *id003
    scheduler: *id004
progress_sample_logger: !new:speechbrain.utils.train_logger.ProgressSampleLogger
  output_path: ./results/tacotron2/9234/samples
  batch_sample_size: 3
  formats:
    raw_batch: raw

text_to_sequence: !name:text_to_sequence.text_to_sequence

pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
  loadables:
    model: !ref <model>
  paths:
    model: Daporte/speechbrain_tacotron2_exp/model.ckpt