Hugging Face's logo

Datasets:
ozguntosun
/
poctts

Dataset card Files Community
ozguntosun commited on
Commit
d41dae2
•
1 Parent(s): 6531e97

Upload args.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. args.json +245 -0
args.json ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_config": "config/vits.json",
3
+ "dataset": [
4
+ "LJSpeech",
5
+ ],
6
+ "dataset_path": {
7
+ "LJSpeech": "wavs",
8
+ },
9
+ "exp_name": "nursubasi",
10
+ "log_dir": "ckpts/tts",
11
+ "model": {
12
+ "filter_channels": 768,
13
+ "gin_channels": 256,
14
+ "hidden_channels": 192,
15
+ "inter_channels": 192,
16
+ "kernel_size": 3,
17
+ "n_heads": 2,
18
+ "n_layers": 6,
19
+ "n_layers_q": 3,
20
+ "n_speakers": 0,
21
+ "p_dropout": 0.1,
22
+ "resblock": "1",
23
+ "resblock_dilation_sizes": [
24
+ [
25
+ 1,
26
+ 3,
27
+ 5,
28
+ ],
29
+ [
30
+ 1,
31
+ 3,
32
+ 5,
33
+ ],
34
+ [
35
+ 1,
36
+ 3,
37
+ 5,
38
+ ],
39
+ ],
40
+ "resblock_kernel_sizes": [
41
+ 3,
42
+ 7,
43
+ 11,
44
+ ],
45
+ "text_token_num": 512,
46
+ "upsample_initial_channel": 512,
47
+ "upsample_kernel_sizes": [
48
+ 16,
49
+ 16,
50
+ 4,
51
+ 4,
52
+ ],
53
+ "upsample_rates": [
54
+ 8,
55
+ 8,
56
+ 2,
57
+ 2,
58
+ ],
59
+ "use_sdp": true,
60
+ "use_spectral_norm": false,
61
+ },
62
+ "model_type": "VITS",
63
+ "preprocess": {
64
+ "align_mel_duration": false,
65
+ "audio_dir": "audios",
66
+ "bits": 8,
67
+ "contentvec_dir": "contentvec",
68
+ "data_augment": false,
69
+ "dur_dir": "durs",
70
+ "duration_dir": "duration",
71
+ "emo2id": "emo2id.json",
72
+ "energy_dir": "energys",
73
+ "energy_extract_mode": "from_mel",
74
+ "energy_norm": false,
75
+ "energy_remove_outlier": false,
76
+ "extract_acoustic_token": false,
77
+ "extract_amplitude_phase": false,
78
+ "extract_audio": true,
79
+ "extract_contentvec_feature": false,
80
+ "extract_duration": false,
81
+ "extract_energy": false,
82
+ "extract_label": false,
83
+ "extract_linear_spec": true,
84
+ "extract_mcep": false,
85
+ "extract_mel": true,
86
+ "extract_mert_feature": false,
87
+ "extract_phone": false,
88
+ "extract_pitch": false,
89
+ "extract_uv": false,
90
+ "extract_wenet_feature": false,
91
+ "extract_whisper_feature": false,
92
+ "file_lst": "file.lst",
93
+ "fmax": null,
94
+ "fmin": 0,
95
+ "hop_size": 256,
96
+ "imaginary_dir": "imaginarys",
97
+ "lab_dir": "labs",
98
+ "label_dir": "labels",
99
+ "language": "tr",
100
+ "lexicon_path": "",
101
+ "linear_dir": "linears",
102
+ "log_amplitude_dir": "log_amplitudes",
103
+ "mcep_dir": "mcep",
104
+ "mel_dir": "mels",
105
+ "mel_extract_mode": "",
106
+ "mel_min_max_norm": false,
107
+ "min_level_db": -115,
108
+ "n_fft": 1024,
109
+ "n_mel": 80,
110
+ "num_silent_frames": 8,
111
+ "phase_dir": "phases",
112
+ "phone_dir": "phones",
113
+ "phone_energy_dir": "phone_energys",
114
+ "phone_extractor": "espeak",
115
+ "phone_pitch_dir": "phone_pitches",
116
+ "phone_seq_file": "phone_seq_file",
117
+ "pitch_dir": "pitches",
118
+ "pitch_extractor": "parselmouth",
119
+ "pitch_norm": false,
120
+ "pitch_remove_outlier": false,
121
+ "processed_dir": "data",
122
+ "raw_data": "raw_data",
123
+ "real_dir": "reals",
124
+ "ref_level_db": 20,
125
+ "sample_rate": 22050,
126
+ "segment_size": 8192,
127
+ "spk2id": "spk2id.json",
128
+ "symbols_dict": "symbols.dict",
129
+ "text_cleaners": [
130
+ "turkish_cleaners",
131
+ ],
132
+ "train_file": "train.json",
133
+ "trim_fft_size": 512,
134
+ "trim_hop_size": 128,
135
+ "trim_silence": false,
136
+ "trim_top_db": 30,
137
+ "trimmed_wav_dir": "trimmed_wavs",
138
+ "use_amplitude_phase": false,
139
+ "use_audio": true,
140
+ "use_dur": false,
141
+ "use_emoid": false,
142
+ "use_frame_duration": false,
143
+ "use_frame_energy": false,
144
+ "use_frame_pitch": false,
145
+ "use_lab": false,
146
+ "use_label": false,
147
+ "use_linear": true,
148
+ "use_log_scale_energy": false,
149
+ "use_log_scale_pitch": false,
150
+ "use_mel": true,
151
+ "use_min_max_norm_mel": false,
152
+ "use_one_hot": false,
153
+ "use_phn_seq": false,
154
+ "use_phone": false,
155
+ "use_phone_duration": false,
156
+ "use_phone_energy": false,
157
+ "use_phone_pitch": false,
158
+ "use_spkid": false,
159
+ "use_text": true,
160
+ "use_uv": false,
161
+ "use_wav": false,
162
+ "use_wenet": false,
163
+ "utt2emo": "utt2emo",
164
+ "utt2spk": "utt2spk",
165
+ "uv_dir": "uvs",
166
+ "valid_file": "test.json",
167
+ "wav_dir": "wavs",
168
+ "wenet_dir": "wenet",
169
+ "win_size": 1024,
170
+ },
171
+ "supported_model_type": [
172
+ "Fastspeech2",
173
+ "VITS",
174
+ "VALLE",
175
+ "NaturalSpeech2",
176
+ ],
177
+ "task_type": "tts",
178
+ "train": {
179
+ "AdamW": {
180
+ "betas": [
181
+ 0.8,
182
+ 0.99,
183
+ ],
184
+ "eps": 1e-09,
185
+ },
186
+ "adamw": {
187
+ "lr": 0.0004,
188
+ },
189
+ "batch_size": 64,
190
+ "betas": [
191
+ 0.8,
192
+ 0.99,
193
+ ],
194
+ "c_kl": 1.0,
195
+ "c_mel": 45,
196
+ "dataloader": {
197
+ "num_worker": 32,
198
+ "pin_memory": true,
199
+ },
200
+ "ddp": true,
201
+ "eps": 1e-09,
202
+ "fp16_run": true,
203
+ "gradient_accumulation_step": 1,
204
+ "init_lr_ratio": 1,
205
+ "keep_checkpoint_max": 5,
206
+ "keep_last": [
207
+ 3,
208
+ -1,
209
+ ],
210
+ "learning_rate": 0.0002,
211
+ "lr_decay": 0.999875,
212
+ "max_epoch": -1,
213
+ "max_steps": 1000000,
214
+ "multi_speaker_training": false,
215
+ "optimizer": "AdamW",
216
+ "random_seed": 10086,
217
+ "reducelronplateau": {
218
+ "factor": 0.8,
219
+ "min_lr": 0.0001,
220
+ "patience": 10,
221
+ },
222
+ "run_eval": [
223
+ false,
224
+ true,
225
+ ],
226
+ "sampler": {
227
+ "drop_last": true,
228
+ "holistic_shuffle": true,
229
+ },
230
+ "save_checkpoint_stride": [
231
+ 5,
232
+ 20,
233
+ ],
234
+ "save_checkpoints_steps": 10000,
235
+ "save_summary_steps": 500,
236
+ "scheduler": "ReduceLROnPlateau",
237
+ "total_training_steps": 50000,
238
+ "tracker": [
239
+ "tensorboard",
240
+ ],
241
+ "valid_interval": 10000,
242
+ "warmup_epochs": 0,
243
+ },
244
+ "use_custom_dataset": [],
245
+ }