-
Notifications
You must be signed in to change notification settings - Fork 1
/
citrinet_256_ru.yaml
472 lines (429 loc) · 14.4 KB
/
citrinet_256_ru.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
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
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
# This config contains the default values for training a Citrinet model with CTC loss and BPE-based vocabulary.
# Default learning parameters in this config are set for effective batch size of 1k on 32 GPUs.
# To train it with smaller batch sizes, you may need to re-tune the learning parameters or use higher accumulate_grad_batches.
# If training for a short time, you can also reduce weight decay to 0.
# Training Recipe
# This model can be trained using the default settings in this config with FP32 precision.
# When training under AMP, increase `warmup_steps` to 5000 for stable training.
# In order to create Citrinet-C, find-replace `filters: 256` with `filters: C`.
# When reducing the receptive field of these models, it is advised to reduce the amount of augmentation
# for larger models from 10x time masking to 5x or 2x time masking.
# For further details regarding Citrinet, visit - https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/asr/configs.html#citrinet
name: &name "Citrinet-256-8x-Stride-ru"
#init_from_model: nemo_experiments/stt_en_citrinet_256-finetuned.nemo # FIXME
init_from_ptl_ckpt: nemo_experiments/stt_en_citrinet_256-finetuned/2022-02-13_17-09-25/checkpoints/stt_en_citrinet_256-finetuned--val_wer=0.5804-epoch=0-last.ckpt
model:
sample_rate: &sample_rate 16000
num_workers: &num_workers 18
labels: &labels
- ' '
- а
- б
- в
- г
- д
- е
- ж
- з
- и
- й
- к
- л
- м
- н
- о
- п
- р
- с
- т
- у
- ф
- х
- ц
- ч
- ш
- щ
- ъ
- ы
- ь
- э
- ю
- я
train_ds:
manifest_filepath:
- data/golos/train/train_all_golos.jsonl
- data/mcv/commonvoice_train_manifest.json
num_workers: *num_workers
sample_rate: 16000
batch_size: 256
trim_silence: false
max_duration: 20
min_duration: 0.1
shuffle: true
is_tarred: false
tarred_audio_filepaths: null
use_start_end_token: false
validation_ds:
manifest_filepath:
- data/golos/test/crowd/test_crowd.jsonl
- data/golos/test/farfield/test_farfield.jsonl
- data/mcv/commonvoice_test_manifest.json
num_workers: *num_workers
sample_rate: 16000
batch_size: 256
shuffle: false
use_start_end_token: false
test_ds:
manifest_filepath: null
sample_rate: 16000
batch_size: 256
shuffle: false
use_start_end_token: false
model_defaults:
repeat: 5
dropout: 0.0
separable: true
se: true
se_context_size: -1
kernel_size_factor: 1.0
tokenizer:
dir: data/an4/tokenizer_spe_unigram_v256 # path to directory which contains either tokenizer.model (bpe) or vocab.txt (for wpe)
type: bpe # Can be either bpe or wpe
preprocessor:
_target_: nemo.collections.asr.modules.AudioToMelSpectrogramPreprocessor
sample_rate: *sample_rate
normalize: "per_feature"
window_size: 0.025
window_stride: 0.01
window: "hann"
features: &n_mels 80
n_fft: 512
frame_splicing: 1
dither: 0.00001
pad_to: 16
stft_conv: false
spec_augment:
_target_: nemo.collections.asr.modules.SpectrogramAugmentation
freq_masks: 2
time_masks: 2
freq_width: 27
time_width: 0.05
encoder:
_target_: nemo.collections.asr.modules.ConvASREncoder
feat_in: *n_mels
activation: relu
conv_mask: true
jasper:
- filters: 256
repeat: 1
kernel: [5]
stride: [1]
dilation: [1]
dropout: 0.0
residual: false
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [11]
stride: [2]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
stride_last: true
residual_mode: "stride_add"
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [13]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [15]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [17]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [19]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [21]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [13]
stride: [2] # *stride
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
stride_last: true
residual_mode: "stride_add"
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [15]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [17]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [19]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [21]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [23]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [25]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [25]
stride: [2] # stride
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
stride_last: true
residual_mode: "stride_add"
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [27]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [29]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [31]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [33]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [35]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [37]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: 256
repeat: ${model.model_defaults.repeat}
kernel: [39]
stride: [1]
dilation: [1]
dropout: ${model.model_defaults.dropout}
residual: true
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
- filters: &enc_final 640
repeat: 1
kernel: [41]
stride: [1]
dilation: [1]
dropout: 0.0
residual: false
separable: ${model.model_defaults.separable}
se: ${model.model_defaults.se}
se_context_size: ${model.model_defaults.se_context_size}
kernel_size_factor: ${model.model_defaults.kernel_size_factor}
decoder:
_target_: nemo.collections.asr.modules.ConvASRDecoder
feat_in: *enc_final
num_classes: 256 # filled with vocabulary size from tokenizer at runtime
optim:
name: novograd
lr: 0.05
# optimizer arguments
betas: [0.8, 0.25]
weight_decay: 0.001
# scheduler setup
sched:
name: CosineAnnealing
# scheduler config override
warmup_steps: 2500
warmup_ratio: null
min_lr: 1e-5
last_epoch: -1
trainer:
gpus: 1 # number of gpus
max_epochs: 100
max_steps: -1
num_nodes: 1
#accelerator: ddp
accumulate_grad_batches: 4
checkpoint_callback: false # Provided by exp_manager
logger: false # Provided by exp_manager
log_every_n_steps: 100 # Interval of logging.
val_check_interval: 0.25
check_val_every_n_epoch: 1
precision: 16
sync_batchnorm: false
benchmark: false
exp_manager:
exp_dir: null
name: *name
create_tensorboard_logger: true
create_checkpoint_callback: true
checkpoint_callback_params:
monitor: "val_wer"
mode: "min"
save_top_k: 3
create_wandb_logger: false
wandb_logger_kwargs:
name: null
project: null
entity: null
resume_if_exists: false
resume_ignore_no_checkpoint: false
hydra:
run:
dir: .
job_logging:
root:
handlers: null