Kamtera commited on
Commit
b872d47
1 Parent(s): b061a6f

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +291 -0
config.json ADDED
@@ -0,0 +1,291 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "output_path": "/kaggle/working/train_output",
3
+ "logger_uri": null,
4
+ "run_name": "vits-male-finetune",
5
+ "project_name": null,
6
+ "run_description": "\ud83d\udc38Coqui trainer run.",
7
+ "print_step": 25,
8
+ "plot_step": 100,
9
+ "model_param_stats": false,
10
+ "wandb_entity": null,
11
+ "dashboard_logger": "tensorboard",
12
+ "log_model_step": null,
13
+ "save_step": 1000,
14
+ "save_n_checkpoints": 5,
15
+ "save_checkpoints": true,
16
+ "save_all_best": false,
17
+ "save_best_after": 10000,
18
+ "target_loss": null,
19
+ "print_eval": true,
20
+ "test_delay_epochs": -1,
21
+ "run_eval": true,
22
+ "run_eval_steps": null,
23
+ "distributed_backend": "nccl",
24
+ "distributed_url": "tcp://localhost:54321",
25
+ "mixed_precision": false,
26
+ "epochs": 1000,
27
+ "batch_size": 8,
28
+ "eval_batch_size": 4,
29
+ "grad_clip": [
30
+ 1000,
31
+ 1000
32
+ ],
33
+ "scheduler_after_epoch": true,
34
+ "lr": 0.001,
35
+ "optimizer": "AdamW",
36
+ "optimizer_params": {
37
+ "betas": [
38
+ 0.8,
39
+ 0.99
40
+ ],
41
+ "eps": 1e-09,
42
+ "weight_decay": 0.01
43
+ },
44
+ "lr_scheduler": null,
45
+ "lr_scheduler_params": null,
46
+ "use_grad_scaler": false,
47
+ "cudnn_enable": true,
48
+ "cudnn_deterministic": false,
49
+ "cudnn_benchmark": false,
50
+ "training_seed": 54321,
51
+ "model": "vits",
52
+ "num_loader_workers": 0,
53
+ "num_eval_loader_workers": 2,
54
+ "use_noise_augment": false,
55
+ "audio": {
56
+ "fft_size": 1024,
57
+ "win_length": 1024,
58
+ "hop_length": 256,
59
+ "frame_shift_ms": null,
60
+ "frame_length_ms": null,
61
+ "stft_pad_mode": "reflect",
62
+ "sample_rate": 22050,
63
+ "resample": false,
64
+ "preemphasis": 0.0,
65
+ "ref_level_db": 20,
66
+ "do_sound_norm": false,
67
+ "log_func": "np.log10",
68
+ "do_trim_silence": true,
69
+ "trim_db": 45,
70
+ "do_rms_norm": false,
71
+ "db_level": null,
72
+ "power": 1.5,
73
+ "griffin_lim_iters": 60,
74
+ "num_mels": 80,
75
+ "mel_fmin": 0,
76
+ "mel_fmax": null,
77
+ "spec_gain": 20,
78
+ "do_amp_to_db_linear": true,
79
+ "do_amp_to_db_mel": true,
80
+ "pitch_fmax": 640.0,
81
+ "pitch_fmin": 1.0,
82
+ "signal_norm": true,
83
+ "min_level_db": -100,
84
+ "symmetric_norm": true,
85
+ "max_norm": 4.0,
86
+ "clip_norm": true,
87
+ "stats_path": null
88
+ },
89
+ "use_phonemes": true,
90
+ "phonemizer": "espeak",
91
+ "phoneme_language": "fa",
92
+ "compute_input_seq_cache": true,
93
+ "text_cleaner": "basic_cleaners",
94
+ "enable_eos_bos_chars": false,
95
+ "test_sentences_file": "",
96
+ "phoneme_cache_path": "/kaggle/working/train_output/phoneme_cache",
97
+ "characters": {
98
+ "characters_class": "TTS.tts.utils.text.characters.IPAPhonemes",
99
+ "vocab_dict": null,
100
+ "pad": "<PAD>",
101
+ "eos": "<EOS>",
102
+ "bos": "<BOS>",
103
+ "blank": "<BLNK>",
104
+ "characters": "\u02c8\u02cc\u02d0\u02d1pbtd\u0288\u0256c\u025fk\u0261q\u0262\u0294\u0274\u014b\u0272\u0273n\u0271m\u0299r\u0280\u2c71\u027e\u027d\u0278\u03b2fv\u03b8\u00f0sz\u0283\u0292\u0282\u0290\u00e7\u029dx\u0263\u03c7\u0281\u0127\u0295h\u0266\u026c\u026e\u028b\u0279\u027bj\u0270l\u026d\u028e\u029faegiouwy\u026a\u028a\u0329\u00e6\u0251\u0254\u0259\u025a\u025b\u025d\u0268\u0303\u0289\u028c\u028d0123456789\"#$%*+/=ABCDEFGHIJKLMNOPRSTUVWXYZ[]^_{}",
105
+ "punctuations": "!(),-.:;? \u0320\u060c\u061b\u061f\u200c<>",
106
+ "phonemes": "\u02c8\u02cc\u02d0\u02d1pbtd\u0288\u0256c\u025fk\u0261q\u0262\u0294\u0274\u014b\u0272\u0273n\u0271m\u0299r\u0280\u2c71\u027e\u027d\u0278\u03b2fv\u03b8\u00f0sz\u0283\u0292\u0282\u0290\u00e7\u029dx\u0263\u03c7\u0281\u0127\u0295h\u0266\u026c\u026e\u028b\u0279\u027bj\u0270l\u026d\u028e\u029faegiouwy\u026a\u028a\u0329\u00e6\u0251\u0254\u0259\u025a\u025b\u025d\u0268\u0303\u0289\u028c\u028d0123456789\"#$%*+/=ABCDEFGHIJKLMNOPRSTUVWXYZ[]^_{}",
107
+ "is_unique": true,
108
+ "is_sorted": true
109
+ },
110
+ "add_blank": true,
111
+ "batch_group_size": 5,
112
+ "loss_masking": null,
113
+ "min_audio_len": 1,
114
+ "max_audio_len": Infinity,
115
+ "min_text_len": 1,
116
+ "max_text_len": Infinity,
117
+ "compute_f0": false,
118
+ "compute_linear_spec": true,
119
+ "precompute_num_workers": 0,
120
+ "start_by_longest": false,
121
+ "shuffle": false,
122
+ "drop_last": false,
123
+ "datasets": [
124
+ {
125
+ "formatter": "mozilla",
126
+ "dataset_name": "",
127
+ "path": "/kaggle/input/persian-tts-dataset",
128
+ "meta_file_train": "metadata.csv",
129
+ "ignored_speakers": null,
130
+ "language": "",
131
+ "phonemizer": "",
132
+ "meta_file_val": "",
133
+ "meta_file_attn_mask": ""
134
+ }
135
+ ],
136
+ "test_sentences": [
137
+ [
138
+ "\u0633\u0644\u0637\u0627\u0646 \u0645\u062d\u0645\u0648\u062f \u062f\u0631 \u0632\u0645\u0633\u062a\u0627\u0646\u06cc \u0633\u062e\u062a \u0628\u0647 \u0637\u0644\u062e\u06a9 \u06af\u0641\u062a \u06a9\u0647: \u0628\u0627 \u0627\u06cc\u0646 \u062c\u0627\u0645\u0647 \u06cc \u06cc\u06a9 \u0644\u0627 \u062f\u0631 \u0627\u06cc\u0646 \u0633\u0631\u0645\u0627 \u0686\u0647 \u0645\u06cc \u06a9\u0646\u06cc "
139
+ ],
140
+ [
141
+ "\u0645\u0631\u062f\u06cc \u0646\u0632\u062f \u0628\u0642\u0627\u0644\u06cc \u0622\u0645\u062f \u0648 \u06af\u0641\u062a \u067e\u06cc\u0627\u0632 \u0647\u0645 \u062f\u0647 \u062a\u0627 \u062f\u0647\u0627\u0646 \u0628\u062f\u0627\u0646 \u062e\u0648 \u0634\u0628\u0648\u06cc \u0633\u0627\u0632\u0645."
142
+ ],
143
+ [
144
+ "\u0627\u0632 \u0645\u0627\u0644 \u062e\u0648\u062f \u067e\u0627\u0631\u0647 \u0627\u06cc \u06af\u0648\u0634\u062a \u0628\u0633\u062a\u0627\u0646 \u0648 \u0632\u06cc\u0631\u0647 \u0628\u0627\u06cc\u06cc \u0645\u0639\u0637\u0651\u0631 \u0628\u0633\u0627\u0632"
145
+ ],
146
+ [
147
+ "\u06cc\u06a9 \u0628\u0627\u0631 \u0647\u0645 \u0627\u0632 \u062c\u0647\u0646\u0645 \u0628\u06af\u0648\u06cc\u06cc\u062f."
148
+ ],
149
+ [
150
+ "\u06cc\u06a9\u06cc \u0627\u0633\u0628\u06cc \u0628\u0647 \u0639\u0627\u0631\u06cc\u062a \u062e\u0648\u0627\u0633\u062a"
151
+ ]
152
+ ],
153
+ "eval_split_max_size": null,
154
+ "eval_split_size": 0.01,
155
+ "use_speaker_weighted_sampler": false,
156
+ "speaker_weighted_sampler_alpha": 1.0,
157
+ "use_language_weighted_sampler": false,
158
+ "language_weighted_sampler_alpha": 1.0,
159
+ "use_length_weighted_sampler": false,
160
+ "length_weighted_sampler_alpha": 1.0,
161
+ "model_args": {
162
+ "num_chars": 156,
163
+ "out_channels": 513,
164
+ "spec_segment_size": 32,
165
+ "hidden_channels": 192,
166
+ "hidden_channels_ffn_text_encoder": 768,
167
+ "num_heads_text_encoder": 2,
168
+ "num_layers_text_encoder": 6,
169
+ "kernel_size_text_encoder": 3,
170
+ "dropout_p_text_encoder": 0.1,
171
+ "dropout_p_duration_predictor": 0.5,
172
+ "kernel_size_posterior_encoder": 5,
173
+ "dilation_rate_posterior_encoder": 1,
174
+ "num_layers_posterior_encoder": 16,
175
+ "kernel_size_flow": 5,
176
+ "dilation_rate_flow": 1,
177
+ "num_layers_flow": 4,
178
+ "resblock_type_decoder": "1",
179
+ "resblock_kernel_sizes_decoder": [
180
+ 3,
181
+ 7,
182
+ 11
183
+ ],
184
+ "resblock_dilation_sizes_decoder": [
185
+ [
186
+ 1,
187
+ 3,
188
+ 5
189
+ ],
190
+ [
191
+ 1,
192
+ 3,
193
+ 5
194
+ ],
195
+ [
196
+ 1,
197
+ 3,
198
+ 5
199
+ ]
200
+ ],
201
+ "upsample_rates_decoder": [
202
+ 8,
203
+ 8,
204
+ 2,
205
+ 2
206
+ ],
207
+ "upsample_initial_channel_decoder": 512,
208
+ "upsample_kernel_sizes_decoder": [
209
+ 16,
210
+ 16,
211
+ 4,
212
+ 4
213
+ ],
214
+ "periods_multi_period_discriminator": [
215
+ 2,
216
+ 3,
217
+ 5,
218
+ 7,
219
+ 11
220
+ ],
221
+ "use_sdp": true,
222
+ "noise_scale": 1.0,
223
+ "inference_noise_scale": 0.667,
224
+ "length_scale": 1,
225
+ "noise_scale_dp": 1.0,
226
+ "inference_noise_scale_dp": 1.0,
227
+ "max_inference_len": null,
228
+ "init_discriminator": true,
229
+ "use_spectral_norm_disriminator": false,
230
+ "use_speaker_embedding": false,
231
+ "num_speakers": 0,
232
+ "speakers_file": null,
233
+ "d_vector_file": null,
234
+ "speaker_embedding_channels": 256,
235
+ "use_d_vector_file": false,
236
+ "d_vector_dim": 0,
237
+ "detach_dp_input": true,
238
+ "use_language_embedding": false,
239
+ "embedded_language_dim": 4,
240
+ "num_languages": 0,
241
+ "language_ids_file": null,
242
+ "use_speaker_encoder_as_loss": false,
243
+ "speaker_encoder_config_path": "",
244
+ "speaker_encoder_model_path": "",
245
+ "condition_dp_on_speaker": true,
246
+ "freeze_encoder": false,
247
+ "freeze_DP": false,
248
+ "freeze_PE": false,
249
+ "freeze_flow_decoder": false,
250
+ "freeze_waveform_decoder": false,
251
+ "encoder_sample_rate": null,
252
+ "interpolate_z": true,
253
+ "reinit_DP": false,
254
+ "reinit_text_encoder": false
255
+ },
256
+ "lr_gen": 0.0002,
257
+ "lr_disc": 0.0002,
258
+ "lr_scheduler_gen": "ExponentialLR",
259
+ "lr_scheduler_gen_params": {
260
+ "gamma": 0.999875,
261
+ "last_epoch": -1
262
+ },
263
+ "lr_scheduler_disc": "ExponentialLR",
264
+ "lr_scheduler_disc_params": {
265
+ "gamma": 0.999875,
266
+ "last_epoch": -1
267
+ },
268
+ "kl_loss_alpha": 1.0,
269
+ "disc_loss_alpha": 1.0,
270
+ "gen_loss_alpha": 1.0,
271
+ "feat_loss_alpha": 1.0,
272
+ "mel_loss_alpha": 45.0,
273
+ "dur_loss_alpha": 1.0,
274
+ "speaker_encoder_loss_alpha": 1.0,
275
+ "return_wav": true,
276
+ "use_weighted_sampler": false,
277
+ "weighted_sampler_attrs": null,
278
+ "weighted_sampler_multipliers": null,
279
+ "r": 1,
280
+ "num_speakers": 0,
281
+ "use_speaker_embedding": false,
282
+ "speakers_file": null,
283
+ "speaker_embedding_channels": 256,
284
+ "language_ids_file": null,
285
+ "use_language_embedding": false,
286
+ "use_d_vector_file": false,
287
+ "d_vector_file": null,
288
+ "d_vector_dim": 0,
289
+ "restore_path": "train_output/kagg/checkpoint_48000.pth",
290
+ "github_branch": "inside_docker"
291
+ }