Upload 55 files
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- assamese/female/model/config.yaml +272 -0
- assamese/female/model/energy_stats.npz +3 -0
- assamese/female/model/feats_stats.npz +3 -0
- assamese/female/model/feats_type +1 -0
- assamese/female/model/model.pth +3 -0
- assamese/female/model/pitch_stats.npz +3 -0
- assamese/male/model/config.yaml +272 -0
- assamese/male/model/energy_stats.npz +3 -0
- assamese/male/model/feats_stats.npz +3 -0
- assamese/male/model/feats_type +1 -0
- assamese/male/model/model.pth +3 -0
- assamese/male/model/pitch_stats.npz +3 -0
- bengali/female/model/config.yaml +274 -0
- bengali/female/model/energy_stats.npz +3 -0
- bengali/female/model/feats_stats.npz +3 -0
- bengali/female/model/feats_type +1 -0
- bengali/female/model/model.pth +3 -0
- bengali/female/model/pitch_stats.npz +3 -0
- bengali/male/model/config.yaml +272 -0
- bengali/male/model/energy_stats.npz +3 -0
- bengali/male/model/feats_stats.npz +3 -0
- bengali/male/model/feats_type +1 -0
- bengali/male/model/model.pth +3 -0
- bengali/male/model/pitch_stats.npz +3 -0
- bodo/female/model/config.yaml +280 -0
- bodo/female/model/energy_stats.npz +3 -0
- bodo/female/model/feats_stats.npz +3 -0
- bodo/female/model/feats_type +1 -0
- bodo/female/model/model.pth +3 -0
- bodo/female/model/pitch_stats.npz +3 -0
- charmap/Text_Cleaning.ipynb +332 -0
- charmap/Text_Cleaning.py +76 -0
- charmap/charmap_Bengali.txt +62 -0
- charmap/charmap_Hindi.txt +81 -0
- charmap/charmap_Hindi.xlsx +0 -0
- charmap/charmap_Malayalam.txt +81 -0
- charmap/charmap_Malayalam.xlsx +0 -0
- charmap/charmap_Marathi.txt +81 -0
- charmap/charmap_Marathi.xlsx +0 -0
- charmap/charmap_Tamil.disabled +76 -0
- charmap/charmap_Tamil.xlsx +0 -0
- charmap/charmap_Telugu.txt +81 -0
- charmap/charmap_Telugu.xlsx +0 -0
- english/female/model/config.yaml +266 -0
- english/female/model/energy_stats.npz +3 -0
- english/female/model/feats_stats.npz +3 -0
- english/female/model/feats_type +1 -0
- english/female/model/model.pth +3 -0
- english/female/model/pitch_stats.npz +3 -0
- english/male/model/config.yaml +265 -0
assamese/female/model/config.yaml
ADDED
|
@@ -0,0 +1,272 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 2 |
+
print_config: false
|
| 3 |
+
log_level: INFO
|
| 4 |
+
dry_run: false
|
| 5 |
+
iterator_type: sequence
|
| 6 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 7 |
+
ngpu: 1
|
| 8 |
+
seed: 0
|
| 9 |
+
num_workers: 1
|
| 10 |
+
num_att_plot: 3
|
| 11 |
+
dist_backend: nccl
|
| 12 |
+
dist_init_method: env://
|
| 13 |
+
dist_world_size: 4
|
| 14 |
+
dist_rank: 0
|
| 15 |
+
local_rank: 0
|
| 16 |
+
dist_master_addr: localhost
|
| 17 |
+
dist_master_port: 60699
|
| 18 |
+
dist_launcher: null
|
| 19 |
+
multiprocessing_distributed: true
|
| 20 |
+
unused_parameters: false
|
| 21 |
+
sharded_ddp: false
|
| 22 |
+
cudnn_enabled: true
|
| 23 |
+
cudnn_benchmark: false
|
| 24 |
+
cudnn_deterministic: true
|
| 25 |
+
collect_stats: false
|
| 26 |
+
write_collected_feats: false
|
| 27 |
+
max_epoch: 1000
|
| 28 |
+
patience: null
|
| 29 |
+
val_scheduler_criterion:
|
| 30 |
+
- valid
|
| 31 |
+
- loss
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
best_model_criterion:
|
| 37 |
+
- - valid
|
| 38 |
+
- loss
|
| 39 |
+
- min
|
| 40 |
+
- - train
|
| 41 |
+
- loss
|
| 42 |
+
- min
|
| 43 |
+
keep_nbest_models: 5
|
| 44 |
+
grad_clip: 1.0
|
| 45 |
+
grad_clip_type: 2.0
|
| 46 |
+
grad_noise: false
|
| 47 |
+
accum_grad: 8
|
| 48 |
+
no_forward_run: false
|
| 49 |
+
resume: true
|
| 50 |
+
train_dtype: float32
|
| 51 |
+
use_amp: false
|
| 52 |
+
log_interval: null
|
| 53 |
+
use_tensorboard: true
|
| 54 |
+
use_wandb: false
|
| 55 |
+
wandb_project: null
|
| 56 |
+
wandb_id: null
|
| 57 |
+
wandb_entity: null
|
| 58 |
+
wandb_name: null
|
| 59 |
+
wandb_model_log_interval: -1
|
| 60 |
+
detect_anomaly: false
|
| 61 |
+
pretrain_path: null
|
| 62 |
+
init_param: []
|
| 63 |
+
ignore_init_mismatch: false
|
| 64 |
+
freeze_param: []
|
| 65 |
+
num_iters_per_epoch: 800
|
| 66 |
+
batch_size: 20
|
| 67 |
+
valid_batch_size: null
|
| 68 |
+
batch_bins: 3000000
|
| 69 |
+
valid_batch_bins: null
|
| 70 |
+
train_shape_file:
|
| 71 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 72 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 73 |
+
valid_shape_file:
|
| 74 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 75 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 76 |
+
batch_type: numel
|
| 77 |
+
valid_batch_type: null
|
| 78 |
+
fold_length:
|
| 79 |
+
- 150
|
| 80 |
+
- 204800
|
| 81 |
+
sort_in_batch: descending
|
| 82 |
+
sort_batch: descending
|
| 83 |
+
multiple_iterator: false
|
| 84 |
+
chunk_length: 500
|
| 85 |
+
chunk_shift_ratio: 0.5
|
| 86 |
+
num_cache_chunks: 1024
|
| 87 |
+
train_data_path_and_name_and_type:
|
| 88 |
+
- - dump/raw/tr_no_dev/text
|
| 89 |
+
- text
|
| 90 |
+
- text
|
| 91 |
+
- - duration_info/tr_no_dev/durations
|
| 92 |
+
- durations
|
| 93 |
+
- text_int
|
| 94 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 95 |
+
- speech
|
| 96 |
+
- sound
|
| 97 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
|
| 98 |
+
- pitch
|
| 99 |
+
- npy
|
| 100 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
|
| 101 |
+
- energy
|
| 102 |
+
- npy
|
| 103 |
+
valid_data_path_and_name_and_type:
|
| 104 |
+
- - dump/raw/dev/text
|
| 105 |
+
- text
|
| 106 |
+
- text
|
| 107 |
+
- - duration_info/dev/durations
|
| 108 |
+
- durations
|
| 109 |
+
- text_int
|
| 110 |
+
- - dump/raw/dev/wav.scp
|
| 111 |
+
- speech
|
| 112 |
+
- sound
|
| 113 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
|
| 114 |
+
- pitch
|
| 115 |
+
- npy
|
| 116 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
|
| 117 |
+
- energy
|
| 118 |
+
- npy
|
| 119 |
+
allow_variable_data_keys: false
|
| 120 |
+
max_cache_size: 0.0
|
| 121 |
+
max_cache_fd: 32
|
| 122 |
+
valid_max_cache_size: null
|
| 123 |
+
optim: adam
|
| 124 |
+
optim_conf:
|
| 125 |
+
lr: 1.0
|
| 126 |
+
scheduler: noamlr
|
| 127 |
+
scheduler_conf:
|
| 128 |
+
model_size: 384
|
| 129 |
+
warmup_steps: 4000
|
| 130 |
+
token_list:
|
| 131 |
+
- <blank>
|
| 132 |
+
- <unk>
|
| 133 |
+
- A
|
| 134 |
+
- a
|
| 135 |
+
- i
|
| 136 |
+
- ','
|
| 137 |
+
- r
|
| 138 |
+
- E
|
| 139 |
+
- k
|
| 140 |
+
- t
|
| 141 |
+
- n
|
| 142 |
+
- l
|
| 143 |
+
- o
|
| 144 |
+
- b
|
| 145 |
+
- u
|
| 146 |
+
- m
|
| 147 |
+
- y
|
| 148 |
+
- $
|
| 149 |
+
- .
|
| 150 |
+
- p
|
| 151 |
+
- h
|
| 152 |
+
- d
|
| 153 |
+
- s
|
| 154 |
+
- I
|
| 155 |
+
- g
|
| 156 |
+
- j
|
| 157 |
+
- ऐ
|
| 158 |
+
- c
|
| 159 |
+
- ट
|
| 160 |
+
- ख
|
| 161 |
+
- C
|
| 162 |
+
- w
|
| 163 |
+
- श
|
| 164 |
+
- M
|
| 165 |
+
- B
|
| 166 |
+
- थ
|
| 167 |
+
- ध
|
| 168 |
+
- ण
|
| 169 |
+
- ष
|
| 170 |
+
- ठ
|
| 171 |
+
- घ
|
| 172 |
+
- U
|
| 173 |
+
- P
|
| 174 |
+
- q
|
| 175 |
+
- ड
|
| 176 |
+
- ङ
|
| 177 |
+
- R
|
| 178 |
+
- औ
|
| 179 |
+
- ञ
|
| 180 |
+
- D
|
| 181 |
+
- ढ
|
| 182 |
+
- <sos/eos>
|
| 183 |
+
odim: null
|
| 184 |
+
model_conf: {}
|
| 185 |
+
use_preprocessor: true
|
| 186 |
+
token_type: char
|
| 187 |
+
bpemodel: null
|
| 188 |
+
non_linguistic_symbols: null
|
| 189 |
+
cleaner: null
|
| 190 |
+
g2p: g2p_en_no_space
|
| 191 |
+
feats_extract: fbank
|
| 192 |
+
feats_extract_conf:
|
| 193 |
+
n_fft: 1024
|
| 194 |
+
hop_length: 256
|
| 195 |
+
win_length: 1024
|
| 196 |
+
fs: 22050
|
| 197 |
+
fmin: 0
|
| 198 |
+
fmax: 8000
|
| 199 |
+
n_mels: 80
|
| 200 |
+
normalize: global_mvn
|
| 201 |
+
normalize_conf:
|
| 202 |
+
stats_file: /speech/arun/released_models/tts/female/assamese/fastspeech2_hs/feats_stats.npz
|
| 203 |
+
tts: fastspeech2
|
| 204 |
+
tts_conf:
|
| 205 |
+
adim: 384
|
| 206 |
+
aheads: 2
|
| 207 |
+
elayers: 4
|
| 208 |
+
eunits: 1536
|
| 209 |
+
dlayers: 4
|
| 210 |
+
dunits: 1536
|
| 211 |
+
positionwise_layer_type: conv1d
|
| 212 |
+
positionwise_conv_kernel_size: 3
|
| 213 |
+
duration_predictor_layers: 2
|
| 214 |
+
duration_predictor_chans: 256
|
| 215 |
+
duration_predictor_kernel_size: 3
|
| 216 |
+
postnet_layers: 5
|
| 217 |
+
postnet_filts: 5
|
| 218 |
+
postnet_chans: 256
|
| 219 |
+
use_masking: true
|
| 220 |
+
use_scaled_pos_enc: true
|
| 221 |
+
encoder_normalize_before: true
|
| 222 |
+
decoder_normalize_before: true
|
| 223 |
+
reduction_factor: 1
|
| 224 |
+
init_type: xavier_uniform
|
| 225 |
+
init_enc_alpha: 1.0
|
| 226 |
+
init_dec_alpha: 1.0
|
| 227 |
+
transformer_enc_dropout_rate: 0.2
|
| 228 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 229 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 230 |
+
transformer_dec_dropout_rate: 0.2
|
| 231 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 232 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 233 |
+
pitch_predictor_layers: 5
|
| 234 |
+
pitch_predictor_chans: 256
|
| 235 |
+
pitch_predictor_kernel_size: 5
|
| 236 |
+
pitch_predictor_dropout: 0.5
|
| 237 |
+
pitch_embed_kernel_size: 1
|
| 238 |
+
pitch_embed_dropout: 0.0
|
| 239 |
+
stop_gradient_from_pitch_predictor: true
|
| 240 |
+
energy_predictor_layers: 2
|
| 241 |
+
energy_predictor_chans: 256
|
| 242 |
+
energy_predictor_kernel_size: 3
|
| 243 |
+
energy_predictor_dropout: 0.5
|
| 244 |
+
energy_embed_kernel_size: 1
|
| 245 |
+
energy_embed_dropout: 0.0
|
| 246 |
+
stop_gradient_from_energy_predictor: false
|
| 247 |
+
pitch_extract: dio
|
| 248 |
+
pitch_extract_conf:
|
| 249 |
+
fs: 22050
|
| 250 |
+
n_fft: 1024
|
| 251 |
+
hop_length: 256
|
| 252 |
+
f0max: 400
|
| 253 |
+
f0min: 80
|
| 254 |
+
reduction_factor: 1
|
| 255 |
+
pitch_normalize: global_mvn
|
| 256 |
+
pitch_normalize_conf:
|
| 257 |
+
stats_file: /speech/arun/released_models/tts/female/assamese/fastspeech2_hs/pitch_stats.npz
|
| 258 |
+
energy_extract: energy
|
| 259 |
+
energy_extract_conf:
|
| 260 |
+
fs: 22050
|
| 261 |
+
n_fft: 1024
|
| 262 |
+
hop_length: 256
|
| 263 |
+
win_length: 1024
|
| 264 |
+
reduction_factor: 1
|
| 265 |
+
energy_normalize: global_mvn
|
| 266 |
+
energy_normalize_conf:
|
| 267 |
+
stats_file: /speech/arun/released_models/tts/female/assamese/fastspeech2_hs/energy_stats.npz
|
| 268 |
+
required:
|
| 269 |
+
- output_dir
|
| 270 |
+
- token_list
|
| 271 |
+
version: 0.10.3a3
|
| 272 |
+
distributed: true
|
assamese/female/model/energy_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80879ba3818e253e2cbfb78176b51de0c247fdb17f07a6b2db730c9d0026f31e
|
| 3 |
+
size 770
|
assamese/female/model/feats_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5be4c43d477c15c04189297ad17d1e1c436f31aed3caff14cacebe4fc13308ab
|
| 3 |
+
size 1402
|
assamese/female/model/feats_type
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
raw
|
assamese/female/model/model.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff9efd82aacade9fcd3fc17e350147d3a272356b1f419872ac9fa525b991b578
|
| 3 |
+
size 148685818
|
assamese/female/model/pitch_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0453690cf54a373b7a39ccbc4557f8487496bca2290c32d684369fe05f4ed4fe
|
| 3 |
+
size 770
|
assamese/male/model/config.yaml
ADDED
|
@@ -0,0 +1,272 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 2 |
+
print_config: false
|
| 3 |
+
log_level: INFO
|
| 4 |
+
dry_run: false
|
| 5 |
+
iterator_type: sequence
|
| 6 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 7 |
+
ngpu: 1
|
| 8 |
+
seed: 0
|
| 9 |
+
num_workers: 1
|
| 10 |
+
num_att_plot: 3
|
| 11 |
+
dist_backend: nccl
|
| 12 |
+
dist_init_method: env://
|
| 13 |
+
dist_world_size: 4
|
| 14 |
+
dist_rank: 0
|
| 15 |
+
local_rank: 0
|
| 16 |
+
dist_master_addr: localhost
|
| 17 |
+
dist_master_port: 33317
|
| 18 |
+
dist_launcher: null
|
| 19 |
+
multiprocessing_distributed: true
|
| 20 |
+
unused_parameters: false
|
| 21 |
+
sharded_ddp: false
|
| 22 |
+
cudnn_enabled: true
|
| 23 |
+
cudnn_benchmark: false
|
| 24 |
+
cudnn_deterministic: true
|
| 25 |
+
collect_stats: false
|
| 26 |
+
write_collected_feats: false
|
| 27 |
+
max_epoch: 1000
|
| 28 |
+
patience: null
|
| 29 |
+
val_scheduler_criterion:
|
| 30 |
+
- valid
|
| 31 |
+
- loss
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
best_model_criterion:
|
| 37 |
+
- - valid
|
| 38 |
+
- loss
|
| 39 |
+
- min
|
| 40 |
+
- - train
|
| 41 |
+
- loss
|
| 42 |
+
- min
|
| 43 |
+
keep_nbest_models: 5
|
| 44 |
+
grad_clip: 1.0
|
| 45 |
+
grad_clip_type: 2.0
|
| 46 |
+
grad_noise: false
|
| 47 |
+
accum_grad: 8
|
| 48 |
+
no_forward_run: false
|
| 49 |
+
resume: true
|
| 50 |
+
train_dtype: float32
|
| 51 |
+
use_amp: false
|
| 52 |
+
log_interval: null
|
| 53 |
+
use_tensorboard: true
|
| 54 |
+
use_wandb: false
|
| 55 |
+
wandb_project: null
|
| 56 |
+
wandb_id: null
|
| 57 |
+
wandb_entity: null
|
| 58 |
+
wandb_name: null
|
| 59 |
+
wandb_model_log_interval: -1
|
| 60 |
+
detect_anomaly: false
|
| 61 |
+
pretrain_path: null
|
| 62 |
+
init_param: []
|
| 63 |
+
ignore_init_mismatch: false
|
| 64 |
+
freeze_param: []
|
| 65 |
+
num_iters_per_epoch: 800
|
| 66 |
+
batch_size: 20
|
| 67 |
+
valid_batch_size: null
|
| 68 |
+
batch_bins: 3000000
|
| 69 |
+
valid_batch_bins: null
|
| 70 |
+
train_shape_file:
|
| 71 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 72 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 73 |
+
valid_shape_file:
|
| 74 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 75 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 76 |
+
batch_type: numel
|
| 77 |
+
valid_batch_type: null
|
| 78 |
+
fold_length:
|
| 79 |
+
- 150
|
| 80 |
+
- 204800
|
| 81 |
+
sort_in_batch: descending
|
| 82 |
+
sort_batch: descending
|
| 83 |
+
multiple_iterator: false
|
| 84 |
+
chunk_length: 500
|
| 85 |
+
chunk_shift_ratio: 0.5
|
| 86 |
+
num_cache_chunks: 1024
|
| 87 |
+
train_data_path_and_name_and_type:
|
| 88 |
+
- - dump/raw/tr_no_dev/text
|
| 89 |
+
- text
|
| 90 |
+
- text
|
| 91 |
+
- - duration_info/tr_no_dev/durations
|
| 92 |
+
- durations
|
| 93 |
+
- text_int
|
| 94 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 95 |
+
- speech
|
| 96 |
+
- sound
|
| 97 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
|
| 98 |
+
- pitch
|
| 99 |
+
- npy
|
| 100 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
|
| 101 |
+
- energy
|
| 102 |
+
- npy
|
| 103 |
+
valid_data_path_and_name_and_type:
|
| 104 |
+
- - dump/raw/dev/text
|
| 105 |
+
- text
|
| 106 |
+
- text
|
| 107 |
+
- - duration_info/dev/durations
|
| 108 |
+
- durations
|
| 109 |
+
- text_int
|
| 110 |
+
- - dump/raw/dev/wav.scp
|
| 111 |
+
- speech
|
| 112 |
+
- sound
|
| 113 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
|
| 114 |
+
- pitch
|
| 115 |
+
- npy
|
| 116 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
|
| 117 |
+
- energy
|
| 118 |
+
- npy
|
| 119 |
+
allow_variable_data_keys: false
|
| 120 |
+
max_cache_size: 0.0
|
| 121 |
+
max_cache_fd: 32
|
| 122 |
+
valid_max_cache_size: null
|
| 123 |
+
optim: adam
|
| 124 |
+
optim_conf:
|
| 125 |
+
lr: 1.0
|
| 126 |
+
scheduler: noamlr
|
| 127 |
+
scheduler_conf:
|
| 128 |
+
model_size: 384
|
| 129 |
+
warmup_steps: 4000
|
| 130 |
+
token_list:
|
| 131 |
+
- <blank>
|
| 132 |
+
- <unk>
|
| 133 |
+
- A
|
| 134 |
+
- a
|
| 135 |
+
- i
|
| 136 |
+
- ','
|
| 137 |
+
- r
|
| 138 |
+
- E
|
| 139 |
+
- k
|
| 140 |
+
- n
|
| 141 |
+
- t
|
| 142 |
+
- l
|
| 143 |
+
- b
|
| 144 |
+
- o
|
| 145 |
+
- u
|
| 146 |
+
- m
|
| 147 |
+
- y
|
| 148 |
+
- $
|
| 149 |
+
- .
|
| 150 |
+
- p
|
| 151 |
+
- h
|
| 152 |
+
- d
|
| 153 |
+
- s
|
| 154 |
+
- I
|
| 155 |
+
- g
|
| 156 |
+
- j
|
| 157 |
+
- ट
|
| 158 |
+
- c
|
| 159 |
+
- ऐ
|
| 160 |
+
- C
|
| 161 |
+
- ख
|
| 162 |
+
- w
|
| 163 |
+
- श
|
| 164 |
+
- B
|
| 165 |
+
- थ
|
| 166 |
+
- M
|
| 167 |
+
- ध
|
| 168 |
+
- ष
|
| 169 |
+
- ण
|
| 170 |
+
- ठ
|
| 171 |
+
- घ
|
| 172 |
+
- q
|
| 173 |
+
- P
|
| 174 |
+
- U
|
| 175 |
+
- ड
|
| 176 |
+
- ङ
|
| 177 |
+
- R
|
| 178 |
+
- औ
|
| 179 |
+
- ञ
|
| 180 |
+
- D
|
| 181 |
+
- ढ
|
| 182 |
+
- <sos/eos>
|
| 183 |
+
odim: null
|
| 184 |
+
model_conf: {}
|
| 185 |
+
use_preprocessor: true
|
| 186 |
+
token_type: char
|
| 187 |
+
bpemodel: null
|
| 188 |
+
non_linguistic_symbols: null
|
| 189 |
+
cleaner: null
|
| 190 |
+
g2p: g2p_en_no_space
|
| 191 |
+
feats_extract: fbank
|
| 192 |
+
feats_extract_conf:
|
| 193 |
+
n_fft: 1024
|
| 194 |
+
hop_length: 256
|
| 195 |
+
win_length: 1024
|
| 196 |
+
fs: 22050
|
| 197 |
+
fmin: 0
|
| 198 |
+
fmax: 8000
|
| 199 |
+
n_mels: 80
|
| 200 |
+
normalize: global_mvn
|
| 201 |
+
normalize_conf:
|
| 202 |
+
stats_file: /speech/arun/released_models/tts/male/assamese/fastspeech2_hs/feats_stats.npz
|
| 203 |
+
tts: fastspeech2
|
| 204 |
+
tts_conf:
|
| 205 |
+
adim: 384
|
| 206 |
+
aheads: 2
|
| 207 |
+
elayers: 4
|
| 208 |
+
eunits: 1536
|
| 209 |
+
dlayers: 4
|
| 210 |
+
dunits: 1536
|
| 211 |
+
positionwise_layer_type: conv1d
|
| 212 |
+
positionwise_conv_kernel_size: 3
|
| 213 |
+
duration_predictor_layers: 2
|
| 214 |
+
duration_predictor_chans: 256
|
| 215 |
+
duration_predictor_kernel_size: 3
|
| 216 |
+
postnet_layers: 5
|
| 217 |
+
postnet_filts: 5
|
| 218 |
+
postnet_chans: 256
|
| 219 |
+
use_masking: true
|
| 220 |
+
use_scaled_pos_enc: true
|
| 221 |
+
encoder_normalize_before: true
|
| 222 |
+
decoder_normalize_before: true
|
| 223 |
+
reduction_factor: 1
|
| 224 |
+
init_type: xavier_uniform
|
| 225 |
+
init_enc_alpha: 1.0
|
| 226 |
+
init_dec_alpha: 1.0
|
| 227 |
+
transformer_enc_dropout_rate: 0.2
|
| 228 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 229 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 230 |
+
transformer_dec_dropout_rate: 0.2
|
| 231 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 232 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 233 |
+
pitch_predictor_layers: 5
|
| 234 |
+
pitch_predictor_chans: 256
|
| 235 |
+
pitch_predictor_kernel_size: 5
|
| 236 |
+
pitch_predictor_dropout: 0.5
|
| 237 |
+
pitch_embed_kernel_size: 1
|
| 238 |
+
pitch_embed_dropout: 0.0
|
| 239 |
+
stop_gradient_from_pitch_predictor: true
|
| 240 |
+
energy_predictor_layers: 2
|
| 241 |
+
energy_predictor_chans: 256
|
| 242 |
+
energy_predictor_kernel_size: 3
|
| 243 |
+
energy_predictor_dropout: 0.5
|
| 244 |
+
energy_embed_kernel_size: 1
|
| 245 |
+
energy_embed_dropout: 0.0
|
| 246 |
+
stop_gradient_from_energy_predictor: false
|
| 247 |
+
pitch_extract: dio
|
| 248 |
+
pitch_extract_conf:
|
| 249 |
+
fs: 22050
|
| 250 |
+
n_fft: 1024
|
| 251 |
+
hop_length: 256
|
| 252 |
+
f0max: 350
|
| 253 |
+
f0min: 40
|
| 254 |
+
reduction_factor: 1
|
| 255 |
+
pitch_normalize: global_mvn
|
| 256 |
+
pitch_normalize_conf:
|
| 257 |
+
stats_file: /speech/arun/released_models/tts/male/assamese/fastspeech2_hs/pitch_stats.npz
|
| 258 |
+
energy_extract: energy
|
| 259 |
+
energy_extract_conf:
|
| 260 |
+
fs: 22050
|
| 261 |
+
n_fft: 1024
|
| 262 |
+
hop_length: 256
|
| 263 |
+
win_length: 1024
|
| 264 |
+
reduction_factor: 1
|
| 265 |
+
energy_normalize: global_mvn
|
| 266 |
+
energy_normalize_conf:
|
| 267 |
+
stats_file: /speech/arun/released_models/tts/male/assamese/fastspeech2_hs/energy_stats.npz
|
| 268 |
+
required:
|
| 269 |
+
- output_dir
|
| 270 |
+
- token_list
|
| 271 |
+
version: 0.10.3a3
|
| 272 |
+
distributed: true
|
assamese/male/model/energy_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66cf788e284146143ddc0d81738e2a4cece116bcfa0ca105a518717fee76f681
|
| 3 |
+
size 770
|
assamese/male/model/feats_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69d6c9208c60aa8fbfce176574858e5fa363ba048fb2418bae339a990592f2c4
|
| 3 |
+
size 1402
|
assamese/male/model/feats_type
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
raw
|
assamese/male/model/model.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61e7dd206b37613770973d840f31906e19e51bb6ac771a40bbdd41afa3bce78a
|
| 3 |
+
size 148685818
|
assamese/male/model/pitch_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:216a0a305e14a24fb5b1a90dc2dd44e31fe4f33efc022d1298bdd7d3890c7f2e
|
| 3 |
+
size 770
|
bengali/female/model/config.yaml
ADDED
|
@@ -0,0 +1,274 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 2 |
+
print_config: false
|
| 3 |
+
log_level: INFO
|
| 4 |
+
dry_run: false
|
| 5 |
+
iterator_type: sequence
|
| 6 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 7 |
+
ngpu: 1
|
| 8 |
+
seed: 0
|
| 9 |
+
num_workers: 1
|
| 10 |
+
num_att_plot: 3
|
| 11 |
+
dist_backend: nccl
|
| 12 |
+
dist_init_method: env://
|
| 13 |
+
dist_world_size: 2
|
| 14 |
+
dist_rank: 0
|
| 15 |
+
local_rank: 0
|
| 16 |
+
dist_master_addr: localhost
|
| 17 |
+
dist_master_port: 51317
|
| 18 |
+
dist_launcher: null
|
| 19 |
+
multiprocessing_distributed: true
|
| 20 |
+
unused_parameters: false
|
| 21 |
+
sharded_ddp: false
|
| 22 |
+
cudnn_enabled: true
|
| 23 |
+
cudnn_benchmark: false
|
| 24 |
+
cudnn_deterministic: true
|
| 25 |
+
collect_stats: false
|
| 26 |
+
write_collected_feats: false
|
| 27 |
+
max_epoch: 1000
|
| 28 |
+
patience: null
|
| 29 |
+
val_scheduler_criterion:
|
| 30 |
+
- valid
|
| 31 |
+
- loss
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
best_model_criterion:
|
| 37 |
+
- - valid
|
| 38 |
+
- loss
|
| 39 |
+
- min
|
| 40 |
+
- - train
|
| 41 |
+
- loss
|
| 42 |
+
- min
|
| 43 |
+
keep_nbest_models: 5
|
| 44 |
+
nbest_averaging_interval: 0
|
| 45 |
+
grad_clip: 1.0
|
| 46 |
+
grad_clip_type: 2.0
|
| 47 |
+
grad_noise: false
|
| 48 |
+
accum_grad: 8
|
| 49 |
+
no_forward_run: false
|
| 50 |
+
resume: true
|
| 51 |
+
train_dtype: float32
|
| 52 |
+
use_amp: false
|
| 53 |
+
log_interval: null
|
| 54 |
+
use_matplotlib: true
|
| 55 |
+
use_tensorboard: true
|
| 56 |
+
use_wandb: false
|
| 57 |
+
wandb_project: null
|
| 58 |
+
wandb_id: null
|
| 59 |
+
wandb_entity: null
|
| 60 |
+
wandb_name: null
|
| 61 |
+
wandb_model_log_interval: -1
|
| 62 |
+
detect_anomaly: false
|
| 63 |
+
pretrain_path: null
|
| 64 |
+
init_param: []
|
| 65 |
+
ignore_init_mismatch: false
|
| 66 |
+
freeze_param: []
|
| 67 |
+
num_iters_per_epoch: 800
|
| 68 |
+
batch_size: 20
|
| 69 |
+
valid_batch_size: null
|
| 70 |
+
batch_bins: 3000000
|
| 71 |
+
valid_batch_bins: null
|
| 72 |
+
train_shape_file:
|
| 73 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 74 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 75 |
+
valid_shape_file:
|
| 76 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 77 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 78 |
+
batch_type: numel
|
| 79 |
+
valid_batch_type: null
|
| 80 |
+
fold_length:
|
| 81 |
+
- 150
|
| 82 |
+
- 204800
|
| 83 |
+
sort_in_batch: descending
|
| 84 |
+
sort_batch: descending
|
| 85 |
+
multiple_iterator: false
|
| 86 |
+
chunk_length: 500
|
| 87 |
+
chunk_shift_ratio: 0.5
|
| 88 |
+
num_cache_chunks: 1024
|
| 89 |
+
train_data_path_and_name_and_type:
|
| 90 |
+
- - dump/raw/tr_no_dev/text
|
| 91 |
+
- text
|
| 92 |
+
- text
|
| 93 |
+
- - duration_info/tr_no_dev/durations
|
| 94 |
+
- durations
|
| 95 |
+
- text_int
|
| 96 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 97 |
+
- speech
|
| 98 |
+
- sound
|
| 99 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
|
| 100 |
+
- pitch
|
| 101 |
+
- npy
|
| 102 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
|
| 103 |
+
- energy
|
| 104 |
+
- npy
|
| 105 |
+
valid_data_path_and_name_and_type:
|
| 106 |
+
- - dump/raw/dev/text
|
| 107 |
+
- text
|
| 108 |
+
- text
|
| 109 |
+
- - duration_info/dev/durations
|
| 110 |
+
- durations
|
| 111 |
+
- text_int
|
| 112 |
+
- - dump/raw/dev/wav.scp
|
| 113 |
+
- speech
|
| 114 |
+
- sound
|
| 115 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
|
| 116 |
+
- pitch
|
| 117 |
+
- npy
|
| 118 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
|
| 119 |
+
- energy
|
| 120 |
+
- npy
|
| 121 |
+
allow_variable_data_keys: false
|
| 122 |
+
max_cache_size: 0.0
|
| 123 |
+
max_cache_fd: 32
|
| 124 |
+
valid_max_cache_size: null
|
| 125 |
+
optim: adam
|
| 126 |
+
optim_conf:
|
| 127 |
+
lr: 1.0
|
| 128 |
+
scheduler: noamlr
|
| 129 |
+
scheduler_conf:
|
| 130 |
+
model_size: 384
|
| 131 |
+
warmup_steps: 4000
|
| 132 |
+
token_list:
|
| 133 |
+
- <blank>
|
| 134 |
+
- <unk>
|
| 135 |
+
- A
|
| 136 |
+
- a
|
| 137 |
+
- E
|
| 138 |
+
- r
|
| 139 |
+
- i
|
| 140 |
+
- n
|
| 141 |
+
- k
|
| 142 |
+
- y
|
| 143 |
+
- t
|
| 144 |
+
- b
|
| 145 |
+
- ','
|
| 146 |
+
- s
|
| 147 |
+
- l
|
| 148 |
+
- m
|
| 149 |
+
- p
|
| 150 |
+
- u
|
| 151 |
+
- o
|
| 152 |
+
- d
|
| 153 |
+
- $
|
| 154 |
+
- .
|
| 155 |
+
- ट
|
| 156 |
+
- j
|
| 157 |
+
- g
|
| 158 |
+
- h
|
| 159 |
+
- श
|
| 160 |
+
- C
|
| 161 |
+
- c
|
| 162 |
+
- I
|
| 163 |
+
- B
|
| 164 |
+
- थ
|
| 165 |
+
- ष
|
| 166 |
+
- ध
|
| 167 |
+
- ड
|
| 168 |
+
- ख
|
| 169 |
+
- ण
|
| 170 |
+
- D
|
| 171 |
+
- P
|
| 172 |
+
- q
|
| 173 |
+
- M
|
| 174 |
+
- ङ
|
| 175 |
+
- U
|
| 176 |
+
- ठ
|
| 177 |
+
- R
|
| 178 |
+
- घ
|
| 179 |
+
- ञ
|
| 180 |
+
- ऐ
|
| 181 |
+
- औ
|
| 182 |
+
- J
|
| 183 |
+
- ढ
|
| 184 |
+
- <sos/eos>
|
| 185 |
+
odim: null
|
| 186 |
+
model_conf: {}
|
| 187 |
+
use_preprocessor: true
|
| 188 |
+
token_type: char
|
| 189 |
+
bpemodel: null
|
| 190 |
+
non_linguistic_symbols: null
|
| 191 |
+
cleaner: null
|
| 192 |
+
g2p: g2p_en_no_space
|
| 193 |
+
feats_extract: fbank
|
| 194 |
+
feats_extract_conf:
|
| 195 |
+
n_fft: 1024
|
| 196 |
+
hop_length: 256
|
| 197 |
+
win_length: 1024
|
| 198 |
+
fs: 22050
|
| 199 |
+
fmin: 0
|
| 200 |
+
fmax: 8000
|
| 201 |
+
n_mels: 80
|
| 202 |
+
normalize: global_mvn
|
| 203 |
+
normalize_conf:
|
| 204 |
+
stats_file: /speech/arun/released_models/tts/female/bengali/fastspeech2_hs/feats_stats.npz
|
| 205 |
+
tts: fastspeech2
|
| 206 |
+
tts_conf:
|
| 207 |
+
adim: 384
|
| 208 |
+
aheads: 2
|
| 209 |
+
elayers: 4
|
| 210 |
+
eunits: 1536
|
| 211 |
+
dlayers: 4
|
| 212 |
+
dunits: 1536
|
| 213 |
+
positionwise_layer_type: conv1d
|
| 214 |
+
positionwise_conv_kernel_size: 3
|
| 215 |
+
duration_predictor_layers: 2
|
| 216 |
+
duration_predictor_chans: 256
|
| 217 |
+
duration_predictor_kernel_size: 3
|
| 218 |
+
postnet_layers: 5
|
| 219 |
+
postnet_filts: 5
|
| 220 |
+
postnet_chans: 256
|
| 221 |
+
use_masking: true
|
| 222 |
+
use_scaled_pos_enc: true
|
| 223 |
+
encoder_normalize_before: true
|
| 224 |
+
decoder_normalize_before: true
|
| 225 |
+
reduction_factor: 1
|
| 226 |
+
init_type: xavier_uniform
|
| 227 |
+
init_enc_alpha: 1.0
|
| 228 |
+
init_dec_alpha: 1.0
|
| 229 |
+
transformer_enc_dropout_rate: 0.2
|
| 230 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 231 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 232 |
+
transformer_dec_dropout_rate: 0.2
|
| 233 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 234 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 235 |
+
pitch_predictor_layers: 5
|
| 236 |
+
pitch_predictor_chans: 256
|
| 237 |
+
pitch_predictor_kernel_size: 5
|
| 238 |
+
pitch_predictor_dropout: 0.5
|
| 239 |
+
pitch_embed_kernel_size: 1
|
| 240 |
+
pitch_embed_dropout: 0.0
|
| 241 |
+
stop_gradient_from_pitch_predictor: true
|
| 242 |
+
energy_predictor_layers: 2
|
| 243 |
+
energy_predictor_chans: 256
|
| 244 |
+
energy_predictor_kernel_size: 3
|
| 245 |
+
energy_predictor_dropout: 0.5
|
| 246 |
+
energy_embed_kernel_size: 1
|
| 247 |
+
energy_embed_dropout: 0.0
|
| 248 |
+
stop_gradient_from_energy_predictor: false
|
| 249 |
+
pitch_extract: dio
|
| 250 |
+
pitch_extract_conf:
|
| 251 |
+
fs: 22050
|
| 252 |
+
n_fft: 1024
|
| 253 |
+
hop_length: 256
|
| 254 |
+
f0max: 400
|
| 255 |
+
f0min: 80
|
| 256 |
+
reduction_factor: 1
|
| 257 |
+
pitch_normalize: global_mvn
|
| 258 |
+
pitch_normalize_conf:
|
| 259 |
+
stats_file: /speech/arun/released_models/tts/female/bengali/fastspeech2_hs/pitch_stats.npz
|
| 260 |
+
energy_extract: energy
|
| 261 |
+
energy_extract_conf:
|
| 262 |
+
fs: 22050
|
| 263 |
+
n_fft: 1024
|
| 264 |
+
hop_length: 256
|
| 265 |
+
win_length: 1024
|
| 266 |
+
reduction_factor: 1
|
| 267 |
+
energy_normalize: global_mvn
|
| 268 |
+
energy_normalize_conf:
|
| 269 |
+
stats_file: /speech/arun/released_models/tts/female/bengali/fastspeech2_hs/energy_stats.npz
|
| 270 |
+
required:
|
| 271 |
+
- output_dir
|
| 272 |
+
- token_list
|
| 273 |
+
version: 0.10.7a1
|
| 274 |
+
distributed: true
|
bengali/female/model/energy_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1fb3a5505b1549cb04e05687fcd81f0ef06361f67c61592097f9628d6fd6aa06
|
| 3 |
+
size 770
|
bengali/female/model/feats_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6299d70bc5b2a185c786cb678d8e526bf41464a32bee9ccf85d7aef27205224b
|
| 3 |
+
size 1402
|
bengali/female/model/feats_type
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
raw
|
bengali/female/model/model.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5eb26e66d94ff5ee854f8f809f7ff8eb9761a3e7afe2742cb00cdd8b5ee4b098
|
| 3 |
+
size 148681929
|
bengali/female/model/pitch_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:90624660ddb2568a44ee68b7887ab4335df6f73a57f38b6d11912ce38a349caf
|
| 3 |
+
size 770
|
bengali/male/model/config.yaml
ADDED
|
@@ -0,0 +1,272 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 2 |
+
print_config: false
|
| 3 |
+
log_level: INFO
|
| 4 |
+
dry_run: false
|
| 5 |
+
iterator_type: sequence
|
| 6 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 7 |
+
ngpu: 1
|
| 8 |
+
seed: 0
|
| 9 |
+
num_workers: 1
|
| 10 |
+
num_att_plot: 3
|
| 11 |
+
dist_backend: nccl
|
| 12 |
+
dist_init_method: env://
|
| 13 |
+
dist_world_size: 4
|
| 14 |
+
dist_rank: 0
|
| 15 |
+
local_rank: 0
|
| 16 |
+
dist_master_addr: localhost
|
| 17 |
+
dist_master_port: 40623
|
| 18 |
+
dist_launcher: null
|
| 19 |
+
multiprocessing_distributed: true
|
| 20 |
+
unused_parameters: false
|
| 21 |
+
sharded_ddp: false
|
| 22 |
+
cudnn_enabled: true
|
| 23 |
+
cudnn_benchmark: false
|
| 24 |
+
cudnn_deterministic: true
|
| 25 |
+
collect_stats: false
|
| 26 |
+
write_collected_feats: false
|
| 27 |
+
max_epoch: 1000
|
| 28 |
+
patience: null
|
| 29 |
+
val_scheduler_criterion:
|
| 30 |
+
- valid
|
| 31 |
+
- loss
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
best_model_criterion:
|
| 37 |
+
- - valid
|
| 38 |
+
- loss
|
| 39 |
+
- min
|
| 40 |
+
- - train
|
| 41 |
+
- loss
|
| 42 |
+
- min
|
| 43 |
+
keep_nbest_models: 5
|
| 44 |
+
grad_clip: 1.0
|
| 45 |
+
grad_clip_type: 2.0
|
| 46 |
+
grad_noise: false
|
| 47 |
+
accum_grad: 8
|
| 48 |
+
no_forward_run: false
|
| 49 |
+
resume: true
|
| 50 |
+
train_dtype: float32
|
| 51 |
+
use_amp: false
|
| 52 |
+
log_interval: null
|
| 53 |
+
use_tensorboard: true
|
| 54 |
+
use_wandb: false
|
| 55 |
+
wandb_project: null
|
| 56 |
+
wandb_id: null
|
| 57 |
+
wandb_entity: null
|
| 58 |
+
wandb_name: null
|
| 59 |
+
wandb_model_log_interval: -1
|
| 60 |
+
detect_anomaly: false
|
| 61 |
+
pretrain_path: null
|
| 62 |
+
init_param: []
|
| 63 |
+
ignore_init_mismatch: false
|
| 64 |
+
freeze_param: []
|
| 65 |
+
num_iters_per_epoch: 800
|
| 66 |
+
batch_size: 20
|
| 67 |
+
valid_batch_size: null
|
| 68 |
+
batch_bins: 3000000
|
| 69 |
+
valid_batch_bins: null
|
| 70 |
+
train_shape_file:
|
| 71 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 72 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 73 |
+
valid_shape_file:
|
| 74 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 75 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 76 |
+
batch_type: numel
|
| 77 |
+
valid_batch_type: null
|
| 78 |
+
fold_length:
|
| 79 |
+
- 150
|
| 80 |
+
- 204800
|
| 81 |
+
sort_in_batch: descending
|
| 82 |
+
sort_batch: descending
|
| 83 |
+
multiple_iterator: false
|
| 84 |
+
chunk_length: 500
|
| 85 |
+
chunk_shift_ratio: 0.5
|
| 86 |
+
num_cache_chunks: 1024
|
| 87 |
+
train_data_path_and_name_and_type:
|
| 88 |
+
- - dump/raw/tr_no_dev/text
|
| 89 |
+
- text
|
| 90 |
+
- text
|
| 91 |
+
- - duration_info/tr_no_dev/durations
|
| 92 |
+
- durations
|
| 93 |
+
- text_int
|
| 94 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 95 |
+
- speech
|
| 96 |
+
- sound
|
| 97 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
|
| 98 |
+
- pitch
|
| 99 |
+
- npy
|
| 100 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
|
| 101 |
+
- energy
|
| 102 |
+
- npy
|
| 103 |
+
valid_data_path_and_name_and_type:
|
| 104 |
+
- - dump/raw/dev/text
|
| 105 |
+
- text
|
| 106 |
+
- text
|
| 107 |
+
- - duration_info/dev/durations
|
| 108 |
+
- durations
|
| 109 |
+
- text_int
|
| 110 |
+
- - dump/raw/dev/wav.scp
|
| 111 |
+
- speech
|
| 112 |
+
- sound
|
| 113 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
|
| 114 |
+
- pitch
|
| 115 |
+
- npy
|
| 116 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
|
| 117 |
+
- energy
|
| 118 |
+
- npy
|
| 119 |
+
allow_variable_data_keys: false
|
| 120 |
+
max_cache_size: 0.0
|
| 121 |
+
max_cache_fd: 32
|
| 122 |
+
valid_max_cache_size: null
|
| 123 |
+
optim: adam
|
| 124 |
+
optim_conf:
|
| 125 |
+
lr: 1.0
|
| 126 |
+
scheduler: noamlr
|
| 127 |
+
scheduler_conf:
|
| 128 |
+
model_size: 384
|
| 129 |
+
warmup_steps: 4000
|
| 130 |
+
token_list:
|
| 131 |
+
- <blank>
|
| 132 |
+
- <unk>
|
| 133 |
+
- A
|
| 134 |
+
- a
|
| 135 |
+
- E
|
| 136 |
+
- r
|
| 137 |
+
- i
|
| 138 |
+
- n
|
| 139 |
+
- k
|
| 140 |
+
- y
|
| 141 |
+
- t
|
| 142 |
+
- b
|
| 143 |
+
- s
|
| 144 |
+
- l
|
| 145 |
+
- m
|
| 146 |
+
- p
|
| 147 |
+
- u
|
| 148 |
+
- o
|
| 149 |
+
- d
|
| 150 |
+
- $
|
| 151 |
+
- .
|
| 152 |
+
- ','
|
| 153 |
+
- ट
|
| 154 |
+
- j
|
| 155 |
+
- g
|
| 156 |
+
- h
|
| 157 |
+
- श
|
| 158 |
+
- C
|
| 159 |
+
- c
|
| 160 |
+
- I
|
| 161 |
+
- B
|
| 162 |
+
- ष
|
| 163 |
+
- थ
|
| 164 |
+
- ध
|
| 165 |
+
- ख
|
| 166 |
+
- ड
|
| 167 |
+
- ण
|
| 168 |
+
- D
|
| 169 |
+
- P
|
| 170 |
+
- q
|
| 171 |
+
- M
|
| 172 |
+
- ङ
|
| 173 |
+
- U
|
| 174 |
+
- ठ
|
| 175 |
+
- R
|
| 176 |
+
- घ
|
| 177 |
+
- ञ
|
| 178 |
+
- ऐ
|
| 179 |
+
- औ
|
| 180 |
+
- J
|
| 181 |
+
- ढ
|
| 182 |
+
- <sos/eos>
|
| 183 |
+
odim: null
|
| 184 |
+
model_conf: {}
|
| 185 |
+
use_preprocessor: true
|
| 186 |
+
token_type: char
|
| 187 |
+
bpemodel: null
|
| 188 |
+
non_linguistic_symbols: null
|
| 189 |
+
cleaner: null
|
| 190 |
+
g2p: g2p_en_no_space
|
| 191 |
+
feats_extract: fbank
|
| 192 |
+
feats_extract_conf:
|
| 193 |
+
n_fft: 1024
|
| 194 |
+
hop_length: 256
|
| 195 |
+
win_length: 1024
|
| 196 |
+
fs: 22050
|
| 197 |
+
fmin: 0
|
| 198 |
+
fmax: 8000
|
| 199 |
+
n_mels: 80
|
| 200 |
+
normalize: global_mvn
|
| 201 |
+
normalize_conf:
|
| 202 |
+
stats_file: /speech/arun/released_models/tts/male/bengali/fastspeech2_hs/feats_stats.npz
|
| 203 |
+
tts: fastspeech2
|
| 204 |
+
tts_conf:
|
| 205 |
+
adim: 384
|
| 206 |
+
aheads: 2
|
| 207 |
+
elayers: 4
|
| 208 |
+
eunits: 1536
|
| 209 |
+
dlayers: 4
|
| 210 |
+
dunits: 1536
|
| 211 |
+
positionwise_layer_type: conv1d
|
| 212 |
+
positionwise_conv_kernel_size: 3
|
| 213 |
+
duration_predictor_layers: 2
|
| 214 |
+
duration_predictor_chans: 256
|
| 215 |
+
duration_predictor_kernel_size: 3
|
| 216 |
+
postnet_layers: 5
|
| 217 |
+
postnet_filts: 5
|
| 218 |
+
postnet_chans: 256
|
| 219 |
+
use_masking: true
|
| 220 |
+
use_scaled_pos_enc: true
|
| 221 |
+
encoder_normalize_before: true
|
| 222 |
+
decoder_normalize_before: true
|
| 223 |
+
reduction_factor: 1
|
| 224 |
+
init_type: xavier_uniform
|
| 225 |
+
init_enc_alpha: 1.0
|
| 226 |
+
init_dec_alpha: 1.0
|
| 227 |
+
transformer_enc_dropout_rate: 0.2
|
| 228 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 229 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 230 |
+
transformer_dec_dropout_rate: 0.2
|
| 231 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 232 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 233 |
+
pitch_predictor_layers: 5
|
| 234 |
+
pitch_predictor_chans: 256
|
| 235 |
+
pitch_predictor_kernel_size: 5
|
| 236 |
+
pitch_predictor_dropout: 0.5
|
| 237 |
+
pitch_embed_kernel_size: 1
|
| 238 |
+
pitch_embed_dropout: 0.0
|
| 239 |
+
stop_gradient_from_pitch_predictor: true
|
| 240 |
+
energy_predictor_layers: 2
|
| 241 |
+
energy_predictor_chans: 256
|
| 242 |
+
energy_predictor_kernel_size: 3
|
| 243 |
+
energy_predictor_dropout: 0.5
|
| 244 |
+
energy_embed_kernel_size: 1
|
| 245 |
+
energy_embed_dropout: 0.0
|
| 246 |
+
stop_gradient_from_energy_predictor: false
|
| 247 |
+
pitch_extract: dio
|
| 248 |
+
pitch_extract_conf:
|
| 249 |
+
fs: 22050
|
| 250 |
+
n_fft: 1024
|
| 251 |
+
hop_length: 256
|
| 252 |
+
f0max: 350
|
| 253 |
+
f0min: 40
|
| 254 |
+
reduction_factor: 1
|
| 255 |
+
pitch_normalize: global_mvn
|
| 256 |
+
pitch_normalize_conf:
|
| 257 |
+
stats_file: /speech/arun/released_models/tts/male/bengali/fastspeech2_hs/pitch_stats.npz
|
| 258 |
+
energy_extract: energy
|
| 259 |
+
energy_extract_conf:
|
| 260 |
+
fs: 22050
|
| 261 |
+
n_fft: 1024
|
| 262 |
+
hop_length: 256
|
| 263 |
+
win_length: 1024
|
| 264 |
+
reduction_factor: 1
|
| 265 |
+
energy_normalize: global_mvn
|
| 266 |
+
energy_normalize_conf:
|
| 267 |
+
stats_file: /speech/arun/released_models/tts/male/bengali/fastspeech2_hs/energy_stats.npz
|
| 268 |
+
required:
|
| 269 |
+
- output_dir
|
| 270 |
+
- token_list
|
| 271 |
+
version: 0.10.3a3
|
| 272 |
+
distributed: true
|
bengali/male/model/energy_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57e315fb35f6df786f93b85135af1d33d3536a963cafc6a253379014bacd86af
|
| 3 |
+
size 770
|
bengali/male/model/feats_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc564e7dbf6feb83ce81a81ddf1e5b88b86f7de313c5b864820db0d35a639f3a
|
| 3 |
+
size 1402
|
bengali/male/model/feats_type
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
raw
|
bengali/male/model/model.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:465c2a5b13bcb4d0b0c33ef947810dade8f264acfe3d673d3582a5ac86e6aff5
|
| 3 |
+
size 148685817
|
bengali/male/model/pitch_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd07268ce13cb540cb0a8ed76dcd9ae7df5251e82940c170a662aad471227dfa
|
| 3 |
+
size 770
|
bodo/female/model/config.yaml
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
accum_grad: 8
|
| 2 |
+
allow_variable_data_keys: false
|
| 3 |
+
batch_bins: 3000000
|
| 4 |
+
batch_size: 20
|
| 5 |
+
batch_type: numel
|
| 6 |
+
best_model_criterion:
|
| 7 |
+
- - valid
|
| 8 |
+
- loss
|
| 9 |
+
- min
|
| 10 |
+
- - train
|
| 11 |
+
- loss
|
| 12 |
+
- min
|
| 13 |
+
bpemodel: null
|
| 14 |
+
chunk_length: 500
|
| 15 |
+
chunk_shift_ratio: 0.5
|
| 16 |
+
cleaner: null
|
| 17 |
+
collect_stats: false
|
| 18 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 19 |
+
cudnn_benchmark: false
|
| 20 |
+
cudnn_deterministic: true
|
| 21 |
+
cudnn_enabled: true
|
| 22 |
+
detect_anomaly: false
|
| 23 |
+
dist_backend: nccl
|
| 24 |
+
dist_init_method: env://
|
| 25 |
+
dist_launcher: null
|
| 26 |
+
dist_master_addr: localhost
|
| 27 |
+
dist_master_port: 37083
|
| 28 |
+
dist_rank: 0
|
| 29 |
+
dist_world_size: 2
|
| 30 |
+
distributed: true
|
| 31 |
+
dry_run: false
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
energy_extract: energy
|
| 37 |
+
energy_extract_conf:
|
| 38 |
+
fs: 22050
|
| 39 |
+
hop_length: 256
|
| 40 |
+
n_fft: 1024
|
| 41 |
+
reduction_factor: 1
|
| 42 |
+
win_length: 1024
|
| 43 |
+
energy_normalize: global_mvn
|
| 44 |
+
energy_normalize_conf:
|
| 45 |
+
stats_file: /home/speech/Fastspeech2_HS/bodo/female/model/energy_stats.npz
|
| 46 |
+
feats_extract: fbank
|
| 47 |
+
feats_extract_conf:
|
| 48 |
+
fmax: 8000
|
| 49 |
+
fmin: 0
|
| 50 |
+
fs: 22050
|
| 51 |
+
hop_length: 256
|
| 52 |
+
n_fft: 1024
|
| 53 |
+
n_mels: 80
|
| 54 |
+
win_length: 1024
|
| 55 |
+
fold_length:
|
| 56 |
+
- 150
|
| 57 |
+
- 204800
|
| 58 |
+
freeze_param: []
|
| 59 |
+
g2p: g2p_en_no_space
|
| 60 |
+
grad_clip: 1.0
|
| 61 |
+
grad_clip_type: 2.0
|
| 62 |
+
grad_noise: false
|
| 63 |
+
ignore_init_mismatch: false
|
| 64 |
+
init_param: []
|
| 65 |
+
iterator_type: sequence
|
| 66 |
+
keep_nbest_models: 5
|
| 67 |
+
local_rank: 0
|
| 68 |
+
log_interval: null
|
| 69 |
+
log_level: INFO
|
| 70 |
+
max_cache_fd: 32
|
| 71 |
+
max_cache_size: 0.0
|
| 72 |
+
max_epoch: 1000
|
| 73 |
+
model_conf: {}
|
| 74 |
+
multiple_iterator: false
|
| 75 |
+
multiprocessing_distributed: true
|
| 76 |
+
nbest_averaging_interval: 0
|
| 77 |
+
ngpu: 1
|
| 78 |
+
no_forward_run: false
|
| 79 |
+
non_linguistic_symbols: null
|
| 80 |
+
normalize: global_mvn
|
| 81 |
+
normalize_conf:
|
| 82 |
+
stats_file: /home/speech/Fastspeech2_HS/bodo/female/model/feats_stats.npz
|
| 83 |
+
num_att_plot: 3
|
| 84 |
+
num_cache_chunks: 1024
|
| 85 |
+
num_iters_per_epoch: 800
|
| 86 |
+
num_workers: 1
|
| 87 |
+
odim: null
|
| 88 |
+
optim: adam
|
| 89 |
+
optim_conf:
|
| 90 |
+
lr: 1.0
|
| 91 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 92 |
+
patience: null
|
| 93 |
+
pitch_extract: dio
|
| 94 |
+
pitch_extract_conf:
|
| 95 |
+
f0max: 400
|
| 96 |
+
f0min: 80
|
| 97 |
+
fs: 22050
|
| 98 |
+
hop_length: 256
|
| 99 |
+
n_fft: 1024
|
| 100 |
+
reduction_factor: 1
|
| 101 |
+
pitch_normalize: global_mvn
|
| 102 |
+
pitch_normalize_conf:
|
| 103 |
+
stats_file: /home/speech/Fastspeech2_HS/bodo/female/model/pitch_stats.npz
|
| 104 |
+
pretrain_path: null
|
| 105 |
+
print_config: false
|
| 106 |
+
required:
|
| 107 |
+
- output_dir
|
| 108 |
+
- token_list
|
| 109 |
+
resume: true
|
| 110 |
+
scheduler: noamlr
|
| 111 |
+
scheduler_conf:
|
| 112 |
+
model_size: 384
|
| 113 |
+
warmup_steps: 4000
|
| 114 |
+
seed: 0
|
| 115 |
+
sharded_ddp: false
|
| 116 |
+
sort_batch: descending
|
| 117 |
+
sort_in_batch: descending
|
| 118 |
+
token_list:
|
| 119 |
+
- <blank>
|
| 120 |
+
- <unk>
|
| 121 |
+
- A
|
| 122 |
+
- n
|
| 123 |
+
- o
|
| 124 |
+
- i
|
| 125 |
+
- b
|
| 126 |
+
- r
|
| 127 |
+
- y
|
| 128 |
+
- q
|
| 129 |
+
- s
|
| 130 |
+
- a
|
| 131 |
+
- m
|
| 132 |
+
- ','
|
| 133 |
+
- g
|
| 134 |
+
- j
|
| 135 |
+
- "\u0916"
|
| 136 |
+
- l
|
| 137 |
+
- d
|
| 138 |
+
- E
|
| 139 |
+
- "\u0925"
|
| 140 |
+
- "\u0910"
|
| 141 |
+
- h
|
| 142 |
+
- u
|
| 143 |
+
- $
|
| 144 |
+
- .
|
| 145 |
+
- w
|
| 146 |
+
- P
|
| 147 |
+
- "\u0914"
|
| 148 |
+
- "\u0919"
|
| 149 |
+
- t
|
| 150 |
+
- k
|
| 151 |
+
- "\u091F"
|
| 152 |
+
- p
|
| 153 |
+
- I
|
| 154 |
+
- "\u0921"
|
| 155 |
+
- U
|
| 156 |
+
- B
|
| 157 |
+
- "\u0927"
|
| 158 |
+
- "\u0937"
|
| 159 |
+
- c
|
| 160 |
+
- "\u0936"
|
| 161 |
+
- "\u0923"
|
| 162 |
+
- H
|
| 163 |
+
- R
|
| 164 |
+
- C
|
| 165 |
+
- "\u0918"
|
| 166 |
+
- "\u0920"
|
| 167 |
+
- "\u0D7D"
|
| 168 |
+
- "\u090D"
|
| 169 |
+
- Y
|
| 170 |
+
- D
|
| 171 |
+
- "\u0911"
|
| 172 |
+
- "\u0928"
|
| 173 |
+
- J
|
| 174 |
+
- z
|
| 175 |
+
- "\u091E"
|
| 176 |
+
- <sos/eos>
|
| 177 |
+
token_type: char
|
| 178 |
+
train_data_path_and_name_and_type:
|
| 179 |
+
- - dump/raw/tr_no_dev/text
|
| 180 |
+
- text
|
| 181 |
+
- text
|
| 182 |
+
- - duration_info/tr_no_dev/durations
|
| 183 |
+
- durations
|
| 184 |
+
- text_int
|
| 185 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 186 |
+
- speech
|
| 187 |
+
- sound
|
| 188 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
|
| 189 |
+
- pitch
|
| 190 |
+
- npy
|
| 191 |
+
- - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
|
| 192 |
+
- energy
|
| 193 |
+
- npy
|
| 194 |
+
train_dtype: float32
|
| 195 |
+
train_shape_file:
|
| 196 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 197 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 198 |
+
tts: fastspeech2
|
| 199 |
+
tts_conf:
|
| 200 |
+
adim: 384
|
| 201 |
+
aheads: 2
|
| 202 |
+
decoder_normalize_before: true
|
| 203 |
+
dlayers: 4
|
| 204 |
+
dunits: 1536
|
| 205 |
+
duration_predictor_chans: 256
|
| 206 |
+
duration_predictor_kernel_size: 3
|
| 207 |
+
duration_predictor_layers: 2
|
| 208 |
+
elayers: 4
|
| 209 |
+
encoder_normalize_before: true
|
| 210 |
+
energy_embed_dropout: 0.0
|
| 211 |
+
energy_embed_kernel_size: 1
|
| 212 |
+
energy_predictor_chans: 256
|
| 213 |
+
energy_predictor_dropout: 0.5
|
| 214 |
+
energy_predictor_kernel_size: 3
|
| 215 |
+
energy_predictor_layers: 2
|
| 216 |
+
eunits: 1536
|
| 217 |
+
init_dec_alpha: 1.0
|
| 218 |
+
init_enc_alpha: 1.0
|
| 219 |
+
init_type: xavier_uniform
|
| 220 |
+
pitch_embed_dropout: 0.0
|
| 221 |
+
pitch_embed_kernel_size: 1
|
| 222 |
+
pitch_predictor_chans: 256
|
| 223 |
+
pitch_predictor_dropout: 0.5
|
| 224 |
+
pitch_predictor_kernel_size: 5
|
| 225 |
+
pitch_predictor_layers: 5
|
| 226 |
+
positionwise_conv_kernel_size: 3
|
| 227 |
+
positionwise_layer_type: conv1d
|
| 228 |
+
postnet_chans: 256
|
| 229 |
+
postnet_filts: 5
|
| 230 |
+
postnet_layers: 5
|
| 231 |
+
reduction_factor: 1
|
| 232 |
+
stop_gradient_from_energy_predictor: false
|
| 233 |
+
stop_gradient_from_pitch_predictor: true
|
| 234 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 235 |
+
transformer_dec_dropout_rate: 0.2
|
| 236 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 237 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 238 |
+
transformer_enc_dropout_rate: 0.2
|
| 239 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 240 |
+
use_masking: true
|
| 241 |
+
use_scaled_pos_enc: true
|
| 242 |
+
unused_parameters: false
|
| 243 |
+
use_amp: false
|
| 244 |
+
use_matplotlib: true
|
| 245 |
+
use_preprocessor: true
|
| 246 |
+
use_tensorboard: true
|
| 247 |
+
use_wandb: false
|
| 248 |
+
val_scheduler_criterion:
|
| 249 |
+
- valid
|
| 250 |
+
- loss
|
| 251 |
+
valid_batch_bins: null
|
| 252 |
+
valid_batch_size: null
|
| 253 |
+
valid_batch_type: null
|
| 254 |
+
valid_data_path_and_name_and_type:
|
| 255 |
+
- - dump/raw/dev/text
|
| 256 |
+
- text
|
| 257 |
+
- text
|
| 258 |
+
- - duration_info/dev/durations
|
| 259 |
+
- durations
|
| 260 |
+
- text_int
|
| 261 |
+
- - dump/raw/dev/wav.scp
|
| 262 |
+
- speech
|
| 263 |
+
- sound
|
| 264 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
|
| 265 |
+
- pitch
|
| 266 |
+
- npy
|
| 267 |
+
- - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
|
| 268 |
+
- energy
|
| 269 |
+
- npy
|
| 270 |
+
valid_max_cache_size: null
|
| 271 |
+
valid_shape_file:
|
| 272 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 273 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 274 |
+
version: 0.10.7a1
|
| 275 |
+
wandb_entity: null
|
| 276 |
+
wandb_id: null
|
| 277 |
+
wandb_model_log_interval: -1
|
| 278 |
+
wandb_name: null
|
| 279 |
+
wandb_project: null
|
| 280 |
+
write_collected_feats: false
|
bodo/female/model/energy_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c61fc80f7e22eb170fefaa3efc4ca1fc347b92153873135cc49f768f9f767983
|
| 3 |
+
size 770
|
bodo/female/model/feats_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5d4aa3b2ba214d6b6674f42fbe6ce06da97c36dcc7249914b2689faffcb0dba
|
| 3 |
+
size 1402
|
bodo/female/model/feats_type
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
raw
|
bodo/female/model/model.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:666bdd54c1ed4551ca7be9f919db21938d783c7bebd7939cd893727282a05d0c
|
| 3 |
+
size 148691145
|
bodo/female/model/pitch_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f9dcb33f0d679fb28039ab78883d34e10b5b22a8699d1ac35c2b48a2a69578b
|
| 3 |
+
size 770
|
charmap/Text_Cleaning.ipynb
ADDED
|
@@ -0,0 +1,332 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [
|
| 3 |
+
{
|
| 4 |
+
"cell_type": "code",
|
| 5 |
+
"execution_count": 2,
|
| 6 |
+
"metadata": {
|
| 7 |
+
"collapsed": true
|
| 8 |
+
},
|
| 9 |
+
"outputs": [],
|
| 10 |
+
"source": [
|
| 11 |
+
"############################################################\n",
|
| 12 |
+
"#Author : Bhagyashree\n",
|
| 13 |
+
"#Date : 1st Sept, 2020\n",
|
| 14 |
+
"#Purpose : Text Cleaning\n",
|
| 15 |
+
"#Input : Text file after timestamp removal\n",
|
| 16 |
+
"#Output : Text file after cleaning data\n",
|
| 17 |
+
"############################################################"
|
| 18 |
+
]
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"cell_type": "code",
|
| 22 |
+
"execution_count": 3,
|
| 23 |
+
"metadata": {
|
| 24 |
+
"collapsed": true
|
| 25 |
+
},
|
| 26 |
+
"outputs": [],
|
| 27 |
+
"source": [
|
| 28 |
+
"import nltk\n",
|
| 29 |
+
"import numpy\n",
|
| 30 |
+
"import xlrd\n",
|
| 31 |
+
"import openpyxl \n",
|
| 32 |
+
"import re"
|
| 33 |
+
]
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"cell_type": "code",
|
| 37 |
+
"execution_count": 21,
|
| 38 |
+
"metadata": {},
|
| 39 |
+
"outputs": [
|
| 40 |
+
{
|
| 41 |
+
"name": "stdout",
|
| 42 |
+
"output_type": "stream",
|
| 43 |
+
"text": [
|
| 44 |
+
"अब हम ऑलट्रेशन ऑफ मेमोरेंडम के बारे में बात करेंगे।\n",
|
| 45 |
+
" मेमोरेंडम के विषयों को बदल दिया जा सकता है।\n",
|
| 46 |
+
"कंपनी अधिनियम 1956 के तहत\n",
|
| 47 |
+
"निम्नलिखित प्रक्रिया के अनुसार\n",
|
| 48 |
+
"यदि कोई कंपनी अपना नाम बदलना चाहती है,\n",
|
| 49 |
+
"तो वे विशेष संकल्प में ऐसा कर सकती है।\n",
|
| 50 |
+
"और केंद्र सरकार की मंजूरी के साथ\n",
|
| 51 |
+
"लिखित में\n",
|
| 52 |
+
"हालाँकि, ऐसी किसी भी मंजूरी की आवश्यकता नहीं होती है\n",
|
| 53 |
+
"केवल कंपनी के नाम में परिवर्तन के लिए\n",
|
| 54 |
+
"बल्कि यह प्रक्रिया\n",
|
| 55 |
+
"पूरी तरह से निजी शब्द को हटाने के लिए होती है।\n",
|
| 56 |
+
"तो इस मामले में\n",
|
| 57 |
+
"एक सार्वजनिक कंपनी को एक निजी कंपनी के संबंध में,\n",
|
| 58 |
+
"या एक निजी कंपनी को एक सार्वजनिक कंपनी में परिवर्तित करने\n",
|
| 59 |
+
"दूसरी जगह पर पंजीकृत कार्यालय का दोबारा परिवर्तन किया जाना है\n",
|
| 60 |
+
"अगर एक ही शहर, कस्बे\n",
|
| 61 |
+
"या गाँव में एक जगह से\n",
|
| 62 |
+
"30 दिनों का नोटिस दिया जाना होता है।\n",
|
| 63 |
+
"यह रजिस्ट्रार द्वारा परिवर्तन की तिथि के बाद\n",
|
| 64 |
+
"होगा जो उसी को दर्ज करेगा।\n",
|
| 65 |
+
"पंजीकृत कार्यालय के परिवर्तन के मामले में\n",
|
| 66 |
+
"जबकि एक ही राज्य में 1 शहर से दूसरे शहर में,\n",
|
| 67 |
+
"विशेष प्रस्ताव पारित किया जाना आवश्यक है।\n",
|
| 68 |
+
"शेयरधारकों की आम बैठक में\n",
|
| 69 |
+
"और इसकी एक प्रति रजिस्ट्रार के पास कार्यालय को\n",
|
| 70 |
+
"30 दिनों के भीतर बदलने के दाखिल करनी होगी।\n",
|
| 71 |
+
"नोटिस देना होगा।\n",
|
| 72 |
+
"रजिस्ट्रार को कार्यालय के नये पते का\n",
|
| 73 |
+
"तब तक परिवर्तन प्रभावी नहीं होगा।\n",
|
| 74 |
+
"जब तक केंद्र सरकार द्वारा प्रस्ताव की पुष्टि नहीं हो जाती\n",
|
| 75 |
+
"कंपनी किसी भी आधार पर\n",
|
| 76 |
+
"उपधारा 1 से\n",
|
| 77 |
+
"उपधारा 7 में धारा 17 तक उल्लेखित है।\n",
|
| 78 |
+
"जैसा कि अधिनियम के बाद ही परिवर्तन प्रभावी होगा।\n",
|
| 79 |
+
"अपने उद्देश्यों में बदलाव कर सकती है\n",
|
| 80 |
+
"सामान्य बैठक में सदस्यों द्वारा\n",
|
| 81 |
+
"कंपनी संशोधन अधिनियम 1996 को मेमोरेंडम ऑफ एसोसिएशन के\n",
|
| 82 |
+
"केंद्र सरकार के प्रतिबंध के ��ाथ समाप्त कर दिया गया है।\n",
|
| 83 |
+
"विशेष संकल्प के अनुरूप अनुमोदित किए जाने\n",
|
| 84 |
+
"ऑब्जेक्ट क्लॉज में परिवर्तन के प्रति शेयर पूंजी में\n",
|
| 85 |
+
"परिवर्तन की प्रक्रिया और ऐसे परिवर्तन करने के अधिकार।\n",
|
| 86 |
+
"समान्य पहलु को आर्टीकल ऑफ एसोसिएशन में रखा गया है।\n",
|
| 87 |
+
"यदि संबंधित विषय के\n",
|
| 88 |
+
" आर्टीकल ऑफ एसोसिएशन में पीछे की प्रक्रिया नहीं दी गई है।\n",
|
| 89 |
+
"कंपनी को पारित करके आर्टीकल ऑफ एसोसिएशन को बदलना होगा।\n",
|
| 90 |
+
"एक विशेष प्रस्ताव\n",
|
| 91 |
+
"अब, हम ऑलट्रेशन ऑफ पर चर्चा करेंगे।\n",
|
| 92 |
+
"कंपनी अधिनियम 2013 के तहत मेमोरेंडम\n",
|
| 93 |
+
"अधिनियम की धारा 16, 17, 18, 19,\n",
|
| 94 |
+
"1956 के कंपनी 21, 23 और 37 के अनुरूप।\n",
|
| 95 |
+
"कंपनी अधिनियम, 2013 की धारा 13 ने\n",
|
| 96 |
+
" ऑलट्रेशन ऑफ मेमोरेंडम के लिए मानदंड निर्धारित किए हैं।\n",
|
| 97 |
+
"यह वही वर्णन करता है कि जैसा कि धारा 61 में दिया गया है।\n",
|
| 98 |
+
"एक कंपनी विशेष संकल्प द्वारा\n",
|
| 99 |
+
"और प्रावधानों को बदल सकती है।\n",
|
| 100 |
+
"निर्दिष्ट प्रक्रिया के अनुपालन से मेमोरेंडम\n",
|
| 101 |
+
"नाम खंड के परिवर्तन के संबंध में,\n",
|
| 102 |
+
"कंपनी अपना नाम बदल सकती है।\n",
|
| 103 |
+
"अनुमोदन के बाद लिखित रूप में केंद्र सरकार द्वारा\n",
|
| 104 |
+
"इसलिए जब कंपनी के नाम में कोई बदलाव किया जाता है,\n",
|
| 105 |
+
"रजिस्टर में दर्ज करेगा।\n",
|
| 106 |
+
" तो रजिस्ट्रार कंपनी के नया नाम\n",
|
| 107 |
+
"पुराने नाम के स्थान पर\n",
|
| 108 |
+
"निगमन का एक नया प्रमाणपत्र जारी करेगा।\n",
|
| 109 |
+
"कंपनी के पंजीकृत कार्यालय के बारे में\n",
|
| 110 |
+
"जैसा कि पहले उल्लेख किया गया है,\n",
|
| 111 |
+
"तब तक कोई प्रभाव नहीं पड़ेगा जब तक कि वह केंद्र सरकार द्वारा अनुमोदित नहीं हो जाता।\n",
|
| 112 |
+
"ऑब्जेक्ट क्लॉज के किसी भी परिवर्तन के संबंध में,\n",
|
| 113 |
+
"किसी कंपनी के मेमोरेंडम के\n",
|
| 114 |
+
"रजिस्ट्रार को दाखिल करने की\n",
|
| 115 |
+
"30 दिनों की अवधि के भीतर\n",
|
| 116 |
+
"तारीख से पंजीकरण को प्रमाणित करना होता है।\n",
|
| 117 |
+
"इस विशेष प्रस्ताव खंड के उप खंड 6 के खंड A के अनुसार\n",
|
| 118 |
+
"चलिए फिर से संक्षेप में\n",
|
| 119 |
+
" ऑलट्रेशन ऑफ मेमोरेंडम के महत्व के बारे में चर्चा करते हैं\n",
|
| 120 |
+
" क्योंकि ये इस मॉड्यूल के अध्ययन का एक बहुत महत्वपूर्ण हिस्सा है।\n",
|
| 121 |
+
"पहला महत्व यह है\n",
|
| 122 |
+
"कि यह एक बहुत ही आवश्यक दस्तावेज है।\n",
|
| 123 |
+
"एक कंपनी के निगमन के लिए\n",
|
| 124 |
+
"दूसरा, यह कंपनी के पंजीकृत कार्यालय को निर्दिष्ट करते हुए\n",
|
| 125 |
+
"रजिस्ट्रार और अदालत के अधिकार क्षेत्र को निर्धारित करता है।\n",
|
| 126 |
+
"तीसरा, यह कंपनी के अधिकारों में उद्देश्यों को\n",
|
| 127 |
+
"जनता की जानकारी के लिए दर्ज करता है।\n",
|
| 128 |
+
"अगला, यह कंपनी को केवल उन कृत्यों को करने के लिए\n",
|
| 129 |
+
"बाध्य करता है जो कंपनी के ऑब्जेक्ट क्लॉज में शामिल हैं।\n",
|
| 130 |
+
"यह कंपनी की अधिकृत कैपिटल\n",
|
| 131 |
+
"और उसके विभाजन को निश्चित राशि के शेयरों में निर्दिष्ट करता है।\n",
|
| 132 |
+
"पर प्रकाश डालता है।\n",
|
| 133 |
+
"यह कंपनी के सदस्यों के लायबिलिटी\n",
|
| 134 |
+
"अंत में, संघ के नियमों को\n",
|
| 135 |
+
"यह एक कंपनी के भी नियंत्रित करता है।\n"
|
| 136 |
+
]
|
| 137 |
+
}
|
| 138 |
+
],
|
| 139 |
+
"source": [
|
| 140 |
+
"file1 = open(\"recent_deliverables_dec2020/Corporate_Law/Hindi/ankita objects 02_Hindi_new.txt\",\"r+\",encoding='utf-8') \n",
|
| 141 |
+
"data = file1.read()\n",
|
| 142 |
+
"print(data)\n",
|
| 143 |
+
"file1.close()"
|
| 144 |
+
]
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"cell_type": "code",
|
| 148 |
+
"execution_count": 22,
|
| 149 |
+
"metadata": {},
|
| 150 |
+
"outputs": [
|
| 151 |
+
{
|
| 152 |
+
"name": "stdout",
|
| 153 |
+
"output_type": "stream",
|
| 154 |
+
"text": [
|
| 155 |
+
"अब हम ऑलट्रेशन ऑफ मेमोरेंडम के बारे में बात करेंगे\n",
|
| 156 |
+
" मेमोरेंडम के विषयों को बदल दिया जा सकता है\n",
|
| 157 |
+
"कंपनी अधिनियम वन नाइन फाइव सिक्स के तहत\n",
|
| 158 |
+
"निम्नलिखित प्रक्रिया के अनुसार\n",
|
| 159 |
+
"यदि कोई कंपनी अपना नाम बदलना चाहती है\n",
|
| 160 |
+
"तो वे विशेष संकल्प में ऐसा कर सकती है\n",
|
| 161 |
+
"और केंद्र सरकार की मंजूरी के साथ\n",
|
| 162 |
+
"लिखित में\n",
|
| 163 |
+
"हालाँकि ऐसी किसी भी मंजूरी की आवश्यकता नहीं होती है\n",
|
| 164 |
+
"केवल कंपनी के नाम में परिवर्तन के लिए\n",
|
| 165 |
+
"बल्कि यह प्रक्रिया\n",
|
| 166 |
+
"पूरी तरह से निजी शब्द को हटाने के लिए होती है\n",
|
| 167 |
+
"तो इस मामले में\n",
|
| 168 |
+
"एक सार्वजनिक कंपनी को एक निजी कंपनी के संबंध में\n",
|
| 169 |
+
"या एक निजी कंपनी को एक सार्वजनिक कंपनी में परिवर्तित करने\n",
|
| 170 |
+
"दूसरी जगह पर पंजीकृत कार्यालय का दोबारा परिवर्तन किया जाना है\n",
|
| 171 |
+
"अगर एक ही शहर कस्बे\n",
|
| 172 |
+
"या गाँव में एक जगह से\n",
|
| 173 |
+
" थ्री ज़ीरो दिनों का नोटिस दिया जाना होता है\n",
|
| 174 |
+
"यह रजिस्ट्रार द्वारा परिवर्तन की तिथि के बाद\n",
|
| 175 |
+
"होगा जो उसी को दर्ज करेगा\n",
|
| 176 |
+
"पंजीकृत कार्यालय के परिवर्तन के मामले में\n",
|
| 177 |
+
"जबकि एक ही राज्य में वन शहर से दूसरे शहर में\n",
|
| 178 |
+
"विशेष प्रस्ताव पारित किया जाना आवश्यक है\n",
|
| 179 |
+
"शेयरधारकों की आम बैठक में\n",
|
| 180 |
+
"और इसकी एक प्रति रजिस्ट्रार के पास कार्यालय को\n",
|
| 181 |
+
" थ्री ज़ीरो दिनों के भीतर बदलने के दाखिल करनी होगी\n",
|
| 182 |
+
"नोटिस देना होगा\n",
|
| 183 |
+
"रजिस्ट्रार को कार्यालय के नये पते का\n",
|
| 184 |
+
"तब तक परिवर्तन प्रभावी नहीं होगा\n",
|
| 185 |
+
"जब तक केंद्र सरकार द्वारा प्रस्ताव की पुष्टि नहीं हो जाती\n",
|
| 186 |
+
"कंपनी किसी भी आधार पर\n",
|
| 187 |
+
"उपधारा वन से\n",
|
| 188 |
+
"उपधारा सेवेन में ध���रा वन सेवेन तक उल्लेखित है\n",
|
| 189 |
+
"जैसा कि अधिनियम के बाद ही परिवर्तन प्रभावी होगा\n",
|
| 190 |
+
"अपने उद्देश्यों में बदलाव कर सकती है\n",
|
| 191 |
+
"सामान्य बैठक में सदस्यों द्वारा\n",
|
| 192 |
+
"कंपनी संशोधन अधिनियम वन नाइन नाइन सिक्स को मेमोरेंडम ऑफ एसोसिएशन के\n",
|
| 193 |
+
"केंद्र सरकार के प्रतिबंध के साथ समाप्त कर दिया गया है\n",
|
| 194 |
+
"विशेष संकल्प के अनुरूप अनुमोदित किए जाने\n",
|
| 195 |
+
"ऑब्जेक्ट क्लॉज में परिवर्तन के प्रति शेयर पूंजी में\n",
|
| 196 |
+
"परिवर्तन की प्रक्रिया और ऐसे परिवर्तन करने के अधिकार\n",
|
| 197 |
+
"समान्य पहलु को आर्टीकल ऑफ एसोसिएशन में रखा गया है\n",
|
| 198 |
+
"यदि संबंधित विषय के\n",
|
| 199 |
+
" आर्टीकल ऑफ एसोसिएशन में पीछे की प्रक्रिया नहीं दी गई है\n",
|
| 200 |
+
"कंपनी को पारित करके आर्टीकल ऑफ एसोसिएशन को बदलना होगा\n",
|
| 201 |
+
"एक विशेष प्रस्ताव\n",
|
| 202 |
+
"अब हम ऑलट्रेशन ऑफ पर चर्चा करेंगे\n",
|
| 203 |
+
"कंपनी अधिनियम टू ज़ीरो वन थ्री के तहत मेमोरेंडम\n",
|
| 204 |
+
"अधिनियम की धारा वन सिक्स वन सेवेन वन ऐइट वन नाइन \n",
|
| 205 |
+
" वन नाइन फाइव सिक्स के कंपनी टू वन टू थ्री और थ्री सेवेन के अनुरूप\n",
|
| 206 |
+
"कंपनी अधिनियम टू ज़ीरो वन थ्री की धारा वन थ्री ने\n",
|
| 207 |
+
" ऑलट्रेशन ऑफ मेमोरेंडम के लिए मानदंड निर्धारित किए हैं\n",
|
| 208 |
+
"यह वही वर्णन करता है कि जैसा कि धारा सिक्स वन में दिया गया है\n",
|
| 209 |
+
"एक कंपनी विशेष संकल्प द्वारा\n",
|
| 210 |
+
"और प्रावधानों को बदल सकती है\n",
|
| 211 |
+
"निर्दिष्ट प्रक्रिया के अनुपालन से मेमोरेंडम\n",
|
| 212 |
+
"नाम खंड के परिवर्तन के संबंध में\n",
|
| 213 |
+
"कंपनी अपना नाम बदल सकती है\n",
|
| 214 |
+
"अनुमोदन के बाद लिखित रूप में केंद्र सरकार द्वारा\n",
|
| 215 |
+
"इसलिए जब कंपनी के नाम में कोई बदलाव किया जाता है\n",
|
| 216 |
+
"रजिस्टर में दर्ज करेगा\n",
|
| 217 |
+
" तो रजिस्ट्रार कंपनी के नया नाम\n",
|
| 218 |
+
"पुराने नाम के स्थान पर\n",
|
| 219 |
+
"निगमन का एक नया प्रमाणपत्र जारी करेगा\n",
|
| 220 |
+
"कंपनी के पंजीकृत कार्यालय के बारे में\n",
|
| 221 |
+
"जैसा कि पहले उल्लेख किया गया है\n",
|
| 222 |
+
"तब तक कोई प्रभाव नहीं पड़ेगा जब तक कि वह केंद्र सरकार द्वारा अनुमोदित नहीं हो जाता\n",
|
| 223 |
+
"ऑब्जेक्ट क्लॉज के किसी भी परिवर्तन के संबंध में\n",
|
| 224 |
+
"किसी कंपनी के मेमोरेंडम के\n",
|
| 225 |
+
"रजिस्ट्रार को दाखिल करने की\n",
|
| 226 |
+
" थ्री ज़ीरो दिनों की अवधि के भीतर\n",
|
| 227 |
+
"तारीख से पंजीकरण को प्रमाणित करना होता है\n",
|
| 228 |
+
"इस विशेष प्रस्ताव खंड के उप खंड सिक्स के खंड ए के अनुसार\n",
|
| 229 |
+
"चलिए फिर से संक्ष���प में\n",
|
| 230 |
+
" ऑलट्रेशन ऑफ मेमोरेंडम के महत्व के बारे में चर्चा करते हैं\n",
|
| 231 |
+
" क्योंकि ये इस मॉड्यूल के अध्ययन का एक बहुत महत्वपूर्ण हिस्सा है\n",
|
| 232 |
+
"पहला महत्व यह है\n",
|
| 233 |
+
"कि यह एक बहुत ही आवश्यक दस्तावेज है\n",
|
| 234 |
+
"एक कंपनी के निगमन के लिए\n",
|
| 235 |
+
"दूसरा यह कंपनी के पंजीकृत कार्यालय को निर्दिष्ट करते हुए\n",
|
| 236 |
+
"रजिस्ट्रार और अदालत के अधिकार क्षेत्र को निर्धारित करता है\n",
|
| 237 |
+
"तीसरा यह कंपनी के अधिकारों में उद्देश्यों को\n",
|
| 238 |
+
"जनता की जानकारी के लिए दर्ज करता है\n",
|
| 239 |
+
"अगला यह कंपनी को केवल उन कृत्यों को करने के लिए\n",
|
| 240 |
+
"बाध्य करता है जो कंपनी के ऑब्जेक्ट क्लॉज में शामिल हैं\n",
|
| 241 |
+
"यह कंपनी की अधिकृत कैपिटल\n",
|
| 242 |
+
"और उसके विभाजन को निश्चित राशि के शेयरों में निर्दिष्ट करता है\n",
|
| 243 |
+
"पर प्रकाश डालता है\n",
|
| 244 |
+
"यह कंपनी के सदस्यों के लायबिलिटी\n",
|
| 245 |
+
"अंत में संघ के नियमों को\n",
|
| 246 |
+
"यह एक कंपनी के भी नियंत्रित करता है\n"
|
| 247 |
+
]
|
| 248 |
+
}
|
| 249 |
+
],
|
| 250 |
+
"source": [
|
| 251 |
+
"\n",
|
| 252 |
+
"wb_obj = openpyxl.load_workbook(\"charmap_v2_new.xlsx\") \n",
|
| 253 |
+
"sheet_obj = wb_obj.active \n",
|
| 254 |
+
"\n",
|
| 255 |
+
"#data = re.sub('[A-Z]*', '',data)\n",
|
| 256 |
+
"#print(data)\n",
|
| 257 |
+
"data = data.replace('?','')\n",
|
| 258 |
+
"data = data.replace(' ',' ')\n",
|
| 259 |
+
"data = data.replace(';','')\n",
|
| 260 |
+
"data = data.replace(')','')\n",
|
| 261 |
+
"data = data.replace('(','')\n",
|
| 262 |
+
"data = data.replace('!','')\n",
|
| 263 |
+
"data = data.replace(' – ',' ')\n",
|
| 264 |
+
"data = data.replace('-',' ')\n",
|
| 265 |
+
"data = data.replace('।','')\n",
|
| 266 |
+
"data = data.replace('&','')\n",
|
| 267 |
+
"data = data.replace('’','')\n",
|
| 268 |
+
"data = data.replace('‘','')\n",
|
| 269 |
+
"data = data.replace(':','')\n",
|
| 270 |
+
"data = data.replace(',','')\n",
|
| 271 |
+
"data = data.replace('/','')\n",
|
| 272 |
+
"data = data.replace(',','')\n",
|
| 273 |
+
"data = data.replace('.','')\n",
|
| 274 |
+
"data = data.replace('|','')\n",
|
| 275 |
+
"m_row = sheet_obj.max_row \n",
|
| 276 |
+
"line = data\n",
|
| 277 |
+
"\n",
|
| 278 |
+
"for i in range(1,m_row+1):\n",
|
| 279 |
+
" num = sheet_obj.cell(row = i, column = 1).value \n",
|
| 280 |
+
" word = sheet_obj.cell(row = i, column = 2).value\n",
|
| 281 |
+
" #print(num)\n",
|
| 282 |
+
" #print(word)\n",
|
| 283 |
+
" line = line.replace(str(num), word)\n",
|
| 284 |
+
"#print(line)\n",
|
| 285 |
+
"#' '.join(line.split())\n",
|
| 286 |
+
"print(line) \n",
|
| 287 |
+
"file1 = open(\"recent_deliverables_dec2020/Corporate_Law/Hindi/ankita objects 02_Hindi.txt\",\"w+\",encoding='utf-8') \n",
|
| 288 |
+
"file1.write(line)\n",
|
| 289 |
+
"file1.close()"
|
| 290 |
+
]
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"cell_type": "code",
|
| 294 |
+
"execution_count": null,
|
| 295 |
+
"metadata": {
|
| 296 |
+
"collapsed": true
|
| 297 |
+
},
|
| 298 |
+
"outputs": [],
|
| 299 |
+
"source": []
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"cell_type": "code",
|
| 303 |
+
"execution_count": null,
|
| 304 |
+
"metadata": {
|
| 305 |
+
"collapsed": true
|
| 306 |
+
},
|
| 307 |
+
"outputs": [],
|
| 308 |
+
"source": []
|
| 309 |
+
}
|
| 310 |
+
],
|
| 311 |
+
"metadata": {
|
| 312 |
+
"kernelspec": {
|
| 313 |
+
"display_name": "Python 3",
|
| 314 |
+
"language": "python",
|
| 315 |
+
"name": "python3"
|
| 316 |
+
},
|
| 317 |
+
"language_info": {
|
| 318 |
+
"codemirror_mode": {
|
| 319 |
+
"name": "ipython",
|
| 320 |
+
"version": 3
|
| 321 |
+
},
|
| 322 |
+
"file_extension": ".py",
|
| 323 |
+
"mimetype": "text/x-python",
|
| 324 |
+
"name": "python",
|
| 325 |
+
"nbconvert_exporter": "python",
|
| 326 |
+
"pygments_lexer": "ipython3",
|
| 327 |
+
"version": "3.6.3"
|
| 328 |
+
}
|
| 329 |
+
},
|
| 330 |
+
"nbformat": 4,
|
| 331 |
+
"nbformat_minor": 2
|
| 332 |
+
}
|
charmap/Text_Cleaning.py
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python
|
| 2 |
+
# coding: utf-8
|
| 3 |
+
|
| 4 |
+
# In[2]:
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
############################################################
|
| 8 |
+
#Author : Bhagyashree
|
| 9 |
+
#Date : 1st Sept, 2020
|
| 10 |
+
#Purpose : Text Cleaning
|
| 11 |
+
#Input : Text file after timestamp removal
|
| 12 |
+
#Output : Text file after cleaning data
|
| 13 |
+
############################################################
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
# In[3]:
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
import nltk
|
| 20 |
+
import numpy
|
| 21 |
+
import xlrd
|
| 22 |
+
import openpyxl
|
| 23 |
+
import re
|
| 24 |
+
import sys
|
| 25 |
+
|
| 26 |
+
# In[21]:
|
| 27 |
+
|
| 28 |
+
#file1 = open("recent_deliverables_dec2020/Corporate_Law/Hindi/ankita objects 02_Hindi_new.txt","r+",encoding='utf-8')
|
| 29 |
+
file1 = open(sys.argv[1],"r+",encoding='utf-8')
|
| 30 |
+
data = file1.read()
|
| 31 |
+
#print(data)
|
| 32 |
+
file1.close()
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
# In[22]:
|
| 36 |
+
|
| 37 |
+
wb_obj = openpyxl.load_workbook(sys.argv[2])
|
| 38 |
+
sheet_obj = wb_obj.active
|
| 39 |
+
|
| 40 |
+
#data = re.sub('[A-Z]*', '',data)
|
| 41 |
+
#print(data)
|
| 42 |
+
data = data.replace('?','')
|
| 43 |
+
data = data.replace(' ',' ')
|
| 44 |
+
data = data.replace(';','')
|
| 45 |
+
data = data.replace(')','')
|
| 46 |
+
data = data.replace('(','')
|
| 47 |
+
data = data.replace('!','')
|
| 48 |
+
data = data.replace(' – ',' ')
|
| 49 |
+
data = data.replace('-',' ')
|
| 50 |
+
data = data.replace('।','')
|
| 51 |
+
data = data.replace('&','')
|
| 52 |
+
data = data.replace('’','')
|
| 53 |
+
data = data.replace('‘','')
|
| 54 |
+
data = data.replace(':','')
|
| 55 |
+
data = data.replace(',','')
|
| 56 |
+
data = data.replace('/','')
|
| 57 |
+
data = data.replace(',','')
|
| 58 |
+
data = data.replace('.','')
|
| 59 |
+
data = data.replace('|','')
|
| 60 |
+
m_row = sheet_obj.max_row
|
| 61 |
+
line = data
|
| 62 |
+
|
| 63 |
+
for i in range(1,m_row+1):
|
| 64 |
+
num = sheet_obj.cell(row = i, column = 1).value
|
| 65 |
+
word = sheet_obj.cell(row = i, column = 2).value
|
| 66 |
+
#print(num)
|
| 67 |
+
#print(word)
|
| 68 |
+
line = line.replace(str(num), word)
|
| 69 |
+
#print(line)
|
| 70 |
+
#' '.join(line.split())
|
| 71 |
+
#print(line)
|
| 72 |
+
file1 = open(sys.argv[3],"w+",encoding='utf-8')
|
| 73 |
+
#file1 = open("recent_deliverables_dec2020/Corporate_Law/Hindi/ankita objects 02_Hindi.txt","w+",encoding='utf-8')
|
| 74 |
+
file1.write(line)
|
| 75 |
+
file1.close()
|
| 76 |
+
|
charmap/charmap_Bengali.txt
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
0 জিরো
|
| 2 |
+
1 ওয়ান
|
| 3 |
+
2 টু
|
| 4 |
+
3 থ্রী
|
| 5 |
+
4 ফোর
|
| 6 |
+
5 ফাইভ
|
| 7 |
+
6 সিক্স
|
| 8 |
+
7 সেবন
|
| 9 |
+
8 এইট
|
| 10 |
+
9 নাইন
|
| 11 |
+
A এ
|
| 12 |
+
a এ
|
| 13 |
+
B বী
|
| 14 |
+
b বী
|
| 15 |
+
C সী
|
| 16 |
+
c সী
|
| 17 |
+
D ডি
|
| 18 |
+
d ডি
|
| 19 |
+
E ই
|
| 20 |
+
e ই
|
| 21 |
+
F এফ
|
| 22 |
+
f এফ
|
| 23 |
+
G জি
|
| 24 |
+
g জি
|
| 25 |
+
H এছ
|
| 26 |
+
h এছ
|
| 27 |
+
I আই
|
| 28 |
+
i আই
|
| 29 |
+
J জে
|
| 30 |
+
j জে
|
| 31 |
+
K কে
|
| 32 |
+
k কে
|
| 33 |
+
L এল
|
| 34 |
+
l এল
|
| 35 |
+
M এম
|
| 36 |
+
m এম
|
| 37 |
+
N এন
|
| 38 |
+
n এন
|
| 39 |
+
O ও
|
| 40 |
+
o ও
|
| 41 |
+
P পি
|
| 42 |
+
p পি
|
| 43 |
+
Q ক্যু
|
| 44 |
+
q ক্যু
|
| 45 |
+
R আর
|
| 46 |
+
r আর
|
| 47 |
+
S এস
|
| 48 |
+
s এস
|
| 49 |
+
T টি
|
| 50 |
+
t টি
|
| 51 |
+
U ইউ
|
| 52 |
+
u ইউ
|
| 53 |
+
V ভি
|
| 54 |
+
v ভি
|
| 55 |
+
W ডাবলু
|
| 56 |
+
w ডাবলু
|
| 57 |
+
X এক্স
|
| 58 |
+
x এক্স
|
| 59 |
+
Y উহাই
|
| 60 |
+
y উহাই
|
| 61 |
+
Z জেডঃ
|
| 62 |
+
z জেডঃ
|
charmap/charmap_Hindi.txt
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
0 ज़ीरो
|
| 2 |
+
1 वन
|
| 3 |
+
2 टू
|
| 4 |
+
3 थ्री
|
| 5 |
+
4 फोर
|
| 6 |
+
5 फाइव
|
| 7 |
+
6 सिक्स
|
| 8 |
+
7 सेवेन
|
| 9 |
+
8 ऐइट
|
| 10 |
+
9 नाइन
|
| 11 |
+
० ज़ीरो
|
| 12 |
+
१ वन
|
| 13 |
+
२ टू
|
| 14 |
+
३ थ्री
|
| 15 |
+
४ फोर
|
| 16 |
+
५ फाइव
|
| 17 |
+
६ सिक्स
|
| 18 |
+
७ सेवेन
|
| 19 |
+
८ ऐइट
|
| 20 |
+
९ नाइन
|
| 21 |
+
A ए
|
| 22 |
+
a ए
|
| 23 |
+
B बी
|
| 24 |
+
b बी
|
| 25 |
+
C सी
|
| 26 |
+
c सी
|
| 27 |
+
D डी
|
| 28 |
+
d डी
|
| 29 |
+
E इ
|
| 30 |
+
e इ
|
| 31 |
+
F एफ
|
| 32 |
+
f एफ
|
| 33 |
+
G जी
|
| 34 |
+
g जी
|
| 35 |
+
H एच
|
| 36 |
+
h एच
|
| 37 |
+
I आई
|
| 38 |
+
i आई
|
| 39 |
+
J जे
|
| 40 |
+
j जे
|
| 41 |
+
K के
|
| 42 |
+
k के
|
| 43 |
+
L एल
|
| 44 |
+
l एल
|
| 45 |
+
M एम
|
| 46 |
+
m एम
|
| 47 |
+
N एन
|
| 48 |
+
n एन
|
| 49 |
+
O ओ
|
| 50 |
+
o ओ
|
| 51 |
+
P पी
|
| 52 |
+
p पी
|
| 53 |
+
Q क्यू
|
| 54 |
+
q क्यू
|
| 55 |
+
R आर
|
| 56 |
+
r आर
|
| 57 |
+
S एस
|
| 58 |
+
s एस
|
| 59 |
+
T टी
|
| 60 |
+
t टी
|
| 61 |
+
U यू
|
| 62 |
+
u यू
|
| 63 |
+
V वी
|
| 64 |
+
v वी
|
| 65 |
+
W डबलु
|
| 66 |
+
w डबलु
|
| 67 |
+
X एक्स
|
| 68 |
+
x एक्स
|
| 69 |
+
Y व्हाई
|
| 70 |
+
y व्हाई
|
| 71 |
+
Z ज़ेड
|
| 72 |
+
z ज़ेड
|
| 73 |
+
α अल्फा
|
| 74 |
+
φ फ़ाई
|
| 75 |
+
+ प्लस
|
| 76 |
+
= इक्वल
|
| 77 |
+
λ लैम्ब्डा
|
| 78 |
+
∅ नॉट
|
| 79 |
+
ϕ फ़ाई
|
| 80 |
+
𝝅 पाई
|
| 81 |
+
⇒ इम्प्लइज
|
charmap/charmap_Hindi.xlsx
ADDED
|
Binary file (10.8 kB). View file
|
|
|
charmap/charmap_Malayalam.txt
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
0 സീറോ
|
| 2 |
+
1 വൺ
|
| 3 |
+
2 ടു
|
| 4 |
+
3 ത്രീ
|
| 5 |
+
4 ഫോർ
|
| 6 |
+
5 ഫൈവ്
|
| 7 |
+
6 സിക്സ്
|
| 8 |
+
7 സെവൻ
|
| 9 |
+
8 എയ്റ്റ്
|
| 10 |
+
9 നൈൻ
|
| 11 |
+
0 സീറോ
|
| 12 |
+
1 വൺ
|
| 13 |
+
2 ടു
|
| 14 |
+
3 ത്രീ
|
| 15 |
+
4 ഫോർ
|
| 16 |
+
5 ഫൈവ്
|
| 17 |
+
6 സിക്സ്
|
| 18 |
+
7 സെവൻ
|
| 19 |
+
8 എയ്റ്റ്
|
| 20 |
+
9 നൈൻ
|
| 21 |
+
A എ
|
| 22 |
+
a എ
|
| 23 |
+
B ബി
|
| 24 |
+
b ബി
|
| 25 |
+
C സി
|
| 26 |
+
c സി
|
| 27 |
+
D ഡി
|
| 28 |
+
d ഡി
|
| 29 |
+
E ഇ
|
| 30 |
+
e ഇ
|
| 31 |
+
F എഫ്
|
| 32 |
+
f എഫ്
|
| 33 |
+
G ജി
|
| 34 |
+
g ജി
|
| 35 |
+
H എഛ്
|
| 36 |
+
h എഛ്
|
| 37 |
+
I ഐ
|
| 38 |
+
i ഐ
|
| 39 |
+
J ജെ
|
| 40 |
+
j ജെ
|
| 41 |
+
K കെ
|
| 42 |
+
k കെ
|
| 43 |
+
L എൽ
|
| 44 |
+
l എൽ
|
| 45 |
+
M എം
|
| 46 |
+
m എം
|
| 47 |
+
N എൻ
|
| 48 |
+
n എൻ
|
| 49 |
+
O ഒ
|
| 50 |
+
o ഒ
|
| 51 |
+
P പി
|
| 52 |
+
p പി
|
| 53 |
+
Q ക്യു
|
| 54 |
+
q ക്യു
|
| 55 |
+
R ആർ
|
| 56 |
+
r ആർ
|
| 57 |
+
S എസ്
|
| 58 |
+
s എസ്
|
| 59 |
+
T ടി
|
| 60 |
+
t ടി
|
| 61 |
+
U യു
|
| 62 |
+
u യു
|
| 63 |
+
V വി
|
| 64 |
+
v വി
|
| 65 |
+
W ഡബ്ല്യൂ
|
| 66 |
+
w ഡബ്ല്യൂ
|
| 67 |
+
X എക്സ്
|
| 68 |
+
x എക്സ്
|
| 69 |
+
Y വൈ
|
| 70 |
+
y വൈ
|
| 71 |
+
Z സെഡ്
|
| 72 |
+
z സെഡ്
|
| 73 |
+
α ആൽഫ
|
| 74 |
+
φ ഫി
|
| 75 |
+
+ പ്ലസ്
|
| 76 |
+
= ഈക്വൽ
|
| 77 |
+
λ ലാംബ്ടാ
|
| 78 |
+
∅ നോട്ട്
|
| 79 |
+
ϕ സൈ
|
| 80 |
+
𝝅 പൈ
|
| 81 |
+
⇒ ഇമ്പ്ലെയ്സ്
|
charmap/charmap_Malayalam.xlsx
ADDED
|
Binary file (13.1 kB). View file
|
|
|
charmap/charmap_Marathi.txt
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
0 ज़ीरो
|
| 2 |
+
1 वन
|
| 3 |
+
2 टू
|
| 4 |
+
3 थ्री
|
| 5 |
+
4 फोर
|
| 6 |
+
5 फाइव
|
| 7 |
+
6 सिक्स
|
| 8 |
+
7 सेवेन
|
| 9 |
+
8 ऐइट
|
| 10 |
+
9 नाइन
|
| 11 |
+
० ज़ीरो
|
| 12 |
+
१ वन
|
| 13 |
+
२ टू
|
| 14 |
+
३ थ्री
|
| 15 |
+
४ फोर
|
| 16 |
+
५ फाइव
|
| 17 |
+
६ सिक्स
|
| 18 |
+
७ सेवेन
|
| 19 |
+
८ ऐइट
|
| 20 |
+
९ नाइन
|
| 21 |
+
A ए
|
| 22 |
+
a ए
|
| 23 |
+
B बी
|
| 24 |
+
b बी
|
| 25 |
+
C सी
|
| 26 |
+
c सी
|
| 27 |
+
D डी
|
| 28 |
+
d डी
|
| 29 |
+
E इ
|
| 30 |
+
e इ
|
| 31 |
+
F एफ
|
| 32 |
+
f एफ
|
| 33 |
+
G जी
|
| 34 |
+
g जी
|
| 35 |
+
H एच
|
| 36 |
+
h एच
|
| 37 |
+
I आई
|
| 38 |
+
i आई
|
| 39 |
+
J जे
|
| 40 |
+
j जे
|
| 41 |
+
K के
|
| 42 |
+
k के
|
| 43 |
+
L एल
|
| 44 |
+
l एल
|
| 45 |
+
M एम
|
| 46 |
+
m एम
|
| 47 |
+
N एन
|
| 48 |
+
n एन
|
| 49 |
+
O ओ
|
| 50 |
+
o ओ
|
| 51 |
+
P पी
|
| 52 |
+
p पी
|
| 53 |
+
Q क्यू
|
| 54 |
+
q क्यू
|
| 55 |
+
R आर
|
| 56 |
+
r आर
|
| 57 |
+
S एस
|
| 58 |
+
s एस
|
| 59 |
+
T टी
|
| 60 |
+
t टी
|
| 61 |
+
U यू
|
| 62 |
+
u यू
|
| 63 |
+
V वी
|
| 64 |
+
v वी
|
| 65 |
+
W डबलु
|
| 66 |
+
w डबलु
|
| 67 |
+
X एक्स
|
| 68 |
+
x एक्स
|
| 69 |
+
Y व्हाई
|
| 70 |
+
y व्हाई
|
| 71 |
+
Z ज़ेड
|
| 72 |
+
z ज़ेड
|
| 73 |
+
α अल्फा
|
| 74 |
+
φ फ़ाई
|
| 75 |
+
+ प्लस
|
| 76 |
+
= इक्वल
|
| 77 |
+
λ लैम्ब्डा
|
| 78 |
+
∅ नॉट
|
| 79 |
+
ϕ फ़ाई
|
| 80 |
+
𝝅 पाई
|
| 81 |
+
⇒ इम्प्लइज
|
charmap/charmap_Marathi.xlsx
ADDED
|
Binary file (10.8 kB). View file
|
|
|
charmap/charmap_Tamil.disabled
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
0 ஸிரோ
|
| 2 |
+
1 ஒன்னு
|
| 3 |
+
2 டூ
|
| 4 |
+
3 த்ரீ
|
| 5 |
+
4 போர்
|
| 6 |
+
5 பைவ்
|
| 7 |
+
6 சீஸ்
|
| 8 |
+
7 செவென்
|
| 9 |
+
8 எயிட்
|
| 10 |
+
9 நயன்
|
| 11 |
+
A எ
|
| 12 |
+
a எ
|
| 13 |
+
B பி
|
| 14 |
+
b பி
|
| 15 |
+
C சி
|
| 16 |
+
c சி
|
| 17 |
+
D டி
|
| 18 |
+
d டி
|
| 19 |
+
E இ
|
| 20 |
+
e இ
|
| 21 |
+
F எப்
|
| 22 |
+
f எப்
|
| 23 |
+
G ஜி
|
| 24 |
+
g ஜி
|
| 25 |
+
H ஹ்
|
| 26 |
+
h ஹ்
|
| 27 |
+
I ஐ
|
| 28 |
+
i ஐ
|
| 29 |
+
J ஜே
|
| 30 |
+
j ஜே
|
| 31 |
+
K கே
|
| 32 |
+
k கே
|
| 33 |
+
L ல்
|
| 34 |
+
l ல்
|
| 35 |
+
M ம்
|
| 36 |
+
m ம்
|
| 37 |
+
N ன்
|
| 38 |
+
n ன்
|
| 39 |
+
O ஓ
|
| 40 |
+
o ஓ
|
| 41 |
+
P பி
|
| 42 |
+
p பி
|
| 43 |
+
Q கியூ
|
| 44 |
+
q கியூ
|
| 45 |
+
R ர்
|
| 46 |
+
r ர்
|
| 47 |
+
S ஸ்
|
| 48 |
+
s ஸ்
|
| 49 |
+
T டீ
|
| 50 |
+
t டீ
|
| 51 |
+
U யூ
|
| 52 |
+
u யூ
|
| 53 |
+
V வி
|
| 54 |
+
v வி
|
| 55 |
+
W டௌபிள்யூ
|
| 56 |
+
w டௌபிள்யூ
|
| 57 |
+
X எஸ்
|
| 58 |
+
x எஸ்
|
| 59 |
+
Y யை
|
| 60 |
+
y யை
|
| 61 |
+
Z செட்
|
| 62 |
+
z செட்
|
| 63 |
+
α ஆல்பா
|
| 64 |
+
φ பிய்
|
| 65 |
+
+ பிளஸ்
|
| 66 |
+
= ஏகுவாள்
|
| 67 |
+
λ லாம்டா
|
| 68 |
+
∅ நோட்
|
| 69 |
+
ϕ பிய்
|
| 70 |
+
𝝅 பை
|
| 71 |
+
⇒ இம்ப்ளிஸ்
|
| 72 |
+
. பாயிண்ட்
|
| 73 |
+
% பெர்ஸண்டாஜ்
|
| 74 |
+
°C டிகிரிசெல்சியஸ்
|
| 75 |
+
θ தீட்டா
|
| 76 |
+
* இண்டூ
|
charmap/charmap_Tamil.xlsx
ADDED
|
Binary file (11.1 kB). View file
|
|
|
charmap/charmap_Telugu.txt
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
0 జీరొ
|
| 2 |
+
1 వన్
|
| 3 |
+
2 టు
|
| 4 |
+
3 త్రీ
|
| 5 |
+
4 ఫోర్
|
| 6 |
+
5 ఫైవ్
|
| 7 |
+
6 సిక్స్
|
| 8 |
+
7 సెవెన్
|
| 9 |
+
8 ఎఇట్
|
| 10 |
+
9 నైన్
|
| 11 |
+
० జీరొ
|
| 12 |
+
१ వన్
|
| 13 |
+
२ టు
|
| 14 |
+
३ త్రీ
|
| 15 |
+
४ ఫోర్
|
| 16 |
+
५ ఫైవ్
|
| 17 |
+
६ సిక్స్
|
| 18 |
+
७ సెవెన్
|
| 19 |
+
८ ఎఇట్
|
| 20 |
+
९ నైన్
|
| 21 |
+
A ఎ
|
| 22 |
+
a ఎ
|
| 23 |
+
B బి
|
| 24 |
+
b బి
|
| 25 |
+
C సి
|
| 26 |
+
c సి
|
| 27 |
+
D డి
|
| 28 |
+
d డి
|
| 29 |
+
E ఇ
|
| 30 |
+
e ఇ
|
| 31 |
+
F ఎఫ్
|
| 32 |
+
f ఎఫ్
|
| 33 |
+
G జి
|
| 34 |
+
g జి
|
| 35 |
+
H హెచ్
|
| 36 |
+
h హెచ్
|
| 37 |
+
I ఐ
|
| 38 |
+
i ఐ
|
| 39 |
+
J జె
|
| 40 |
+
j జె
|
| 41 |
+
K కె
|
| 42 |
+
k కె
|
| 43 |
+
L ఎల్
|
| 44 |
+
l ఎల్
|
| 45 |
+
M ఎమ్
|
| 46 |
+
m ఎమ్
|
| 47 |
+
N ఎన్
|
| 48 |
+
n ఎన్
|
| 49 |
+
O ఒ
|
| 50 |
+
o ఒ
|
| 51 |
+
P పి
|
| 52 |
+
p పి
|
| 53 |
+
Q క్యు
|
| 54 |
+
q క్యు
|
| 55 |
+
R ఆర్
|
| 56 |
+
r ఆర్
|
| 57 |
+
S ఎస్
|
| 58 |
+
s ఎస్
|
| 59 |
+
T టి
|
| 60 |
+
t టి
|
| 61 |
+
U యు
|
| 62 |
+
u యు
|
| 63 |
+
V వి
|
| 64 |
+
v వి
|
| 65 |
+
W డబ్లు
|
| 66 |
+
w డబ్లు
|
| 67 |
+
X ఎక్స్
|
| 68 |
+
x ఎక్స్
|
| 69 |
+
Y వై
|
| 70 |
+
y వై
|
| 71 |
+
Z జెడ్
|
| 72 |
+
z జెడ్
|
| 73 |
+
α ఆల్ఫ
|
| 74 |
+
φ ఫై
|
| 75 |
+
+ ప్లస్
|
| 76 |
+
= ఈక్వల్
|
| 77 |
+
λ లామ్బ్డ
|
| 78 |
+
∅ నల్
|
| 79 |
+
ϕ ఫై
|
| 80 |
+
𝝅 పై
|
| 81 |
+
⇒ ఇమ్ప్లైస్
|
charmap/charmap_Telugu.xlsx
ADDED
|
Binary file (10 kB). View file
|
|
|
english/female/model/config.yaml
ADDED
|
@@ -0,0 +1,266 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
accum_grad: 8
|
| 2 |
+
allow_variable_data_keys: false
|
| 3 |
+
batch_bins: 3000000
|
| 4 |
+
batch_size: 20
|
| 5 |
+
batch_type: numel
|
| 6 |
+
best_model_criterion:
|
| 7 |
+
- - valid
|
| 8 |
+
- loss
|
| 9 |
+
- min
|
| 10 |
+
- - train
|
| 11 |
+
- loss
|
| 12 |
+
- min
|
| 13 |
+
bpemodel: null
|
| 14 |
+
chunk_length: 500
|
| 15 |
+
chunk_shift_ratio: 0.5
|
| 16 |
+
cleaner: null
|
| 17 |
+
collect_stats: false
|
| 18 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 19 |
+
cudnn_benchmark: false
|
| 20 |
+
cudnn_deterministic: true
|
| 21 |
+
cudnn_enabled: true
|
| 22 |
+
detect_anomaly: false
|
| 23 |
+
dist_backend: nccl
|
| 24 |
+
dist_init_method: env://
|
| 25 |
+
dist_launcher: null
|
| 26 |
+
dist_master_addr: localhost
|
| 27 |
+
dist_master_port: 44035
|
| 28 |
+
dist_rank: 0
|
| 29 |
+
dist_world_size: 8
|
| 30 |
+
distributed: true
|
| 31 |
+
dry_run: false
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
energy_extract: energy
|
| 37 |
+
energy_extract_conf:
|
| 38 |
+
fs: 22050
|
| 39 |
+
hop_length: 256
|
| 40 |
+
n_fft: 1024
|
| 41 |
+
reduction_factor: 1
|
| 42 |
+
win_length: null
|
| 43 |
+
energy_normalize: global_mvn
|
| 44 |
+
energy_normalize_conf:
|
| 45 |
+
stats_file: /home/speech/Fastspeech2_HS/english/female/model/energy_stats.npz
|
| 46 |
+
feats_extract: fbank
|
| 47 |
+
feats_extract_conf:
|
| 48 |
+
fmax: 8000
|
| 49 |
+
fmin: 0
|
| 50 |
+
fs: 22050
|
| 51 |
+
hop_length: 256
|
| 52 |
+
n_fft: 1024
|
| 53 |
+
n_mels: 80
|
| 54 |
+
win_length: null
|
| 55 |
+
fold_length:
|
| 56 |
+
- 150
|
| 57 |
+
- 204800
|
| 58 |
+
freeze_param: []
|
| 59 |
+
g2p: g2p_en_no_space
|
| 60 |
+
grad_clip: 1.0
|
| 61 |
+
grad_clip_type: 2.0
|
| 62 |
+
grad_noise: false
|
| 63 |
+
ignore_init_mismatch: false
|
| 64 |
+
init_param: []
|
| 65 |
+
iterator_type: sequence
|
| 66 |
+
keep_nbest_models: 5
|
| 67 |
+
local_rank: 0
|
| 68 |
+
log_interval: null
|
| 69 |
+
log_level: INFO
|
| 70 |
+
max_cache_fd: 32
|
| 71 |
+
max_cache_size: 0.0
|
| 72 |
+
max_epoch: 1000
|
| 73 |
+
model_conf: {}
|
| 74 |
+
multiple_iterator: false
|
| 75 |
+
multiprocessing_distributed: true
|
| 76 |
+
ngpu: 1
|
| 77 |
+
no_forward_run: false
|
| 78 |
+
non_linguistic_symbols: null
|
| 79 |
+
normalize: global_mvn
|
| 80 |
+
normalize_conf:
|
| 81 |
+
stats_file: /home/speech/Fastspeech2_HS/english/female/model/feats_stats.npz
|
| 82 |
+
num_att_plot: 3
|
| 83 |
+
num_cache_chunks: 1024
|
| 84 |
+
num_iters_per_epoch: 800
|
| 85 |
+
num_workers: 1
|
| 86 |
+
odim: null
|
| 87 |
+
optim: adam
|
| 88 |
+
optim_conf:
|
| 89 |
+
lr: 1.0
|
| 90 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 91 |
+
patience: null
|
| 92 |
+
pitch_extract: dio
|
| 93 |
+
pitch_extract_conf:
|
| 94 |
+
f0max: 400
|
| 95 |
+
f0min: 80
|
| 96 |
+
fs: 22050
|
| 97 |
+
hop_length: 256
|
| 98 |
+
n_fft: 1024
|
| 99 |
+
reduction_factor: 1
|
| 100 |
+
pitch_normalize: global_mvn
|
| 101 |
+
pitch_normalize_conf:
|
| 102 |
+
stats_file: /home/speech/Fastspeech2_HS/english/female/model/pitch_stats.npz
|
| 103 |
+
pretrain_path: null
|
| 104 |
+
print_config: false
|
| 105 |
+
required:
|
| 106 |
+
- output_dir
|
| 107 |
+
- token_list
|
| 108 |
+
resume: true
|
| 109 |
+
scheduler: noamlr
|
| 110 |
+
scheduler_conf:
|
| 111 |
+
model_size: 384
|
| 112 |
+
warmup_steps: 4000
|
| 113 |
+
seed: 0
|
| 114 |
+
sharded_ddp: false
|
| 115 |
+
sort_batch: descending
|
| 116 |
+
sort_in_batch: descending
|
| 117 |
+
token_list:
|
| 118 |
+
- <blank>
|
| 119 |
+
- <unk>
|
| 120 |
+
- <space>
|
| 121 |
+
- a
|
| 122 |
+
- r
|
| 123 |
+
- n
|
| 124 |
+
- "\u091F"
|
| 125 |
+
- i
|
| 126 |
+
- "\u0921"
|
| 127 |
+
- E
|
| 128 |
+
- s
|
| 129 |
+
- l
|
| 130 |
+
- d
|
| 131 |
+
- w
|
| 132 |
+
- I
|
| 133 |
+
- m
|
| 134 |
+
- k
|
| 135 |
+
- z
|
| 136 |
+
- "\u0905"
|
| 137 |
+
- f
|
| 138 |
+
- h
|
| 139 |
+
- "\u0911"
|
| 140 |
+
- U
|
| 141 |
+
- A
|
| 142 |
+
- .
|
| 143 |
+
- "\u0910"
|
| 144 |
+
- b
|
| 145 |
+
- p
|
| 146 |
+
- ','
|
| 147 |
+
- "\u0919"
|
| 148 |
+
- o
|
| 149 |
+
- g
|
| 150 |
+
- y
|
| 151 |
+
- "\u0936"
|
| 152 |
+
- "\u0914"
|
| 153 |
+
- t
|
| 154 |
+
- u
|
| 155 |
+
- j
|
| 156 |
+
- c
|
| 157 |
+
- '?'
|
| 158 |
+
- '!'
|
| 159 |
+
- q
|
| 160 |
+
- "\u0923"
|
| 161 |
+
- "\u0925"
|
| 162 |
+
- "\u0937"
|
| 163 |
+
- "\u0927"
|
| 164 |
+
- B
|
| 165 |
+
- H
|
| 166 |
+
- P
|
| 167 |
+
- D
|
| 168 |
+
- M
|
| 169 |
+
- C
|
| 170 |
+
- R
|
| 171 |
+
- "\u0918"
|
| 172 |
+
- "\u0916"
|
| 173 |
+
- O
|
| 174 |
+
- v
|
| 175 |
+
- <sos/eos>
|
| 176 |
+
token_type: char
|
| 177 |
+
train_data_path_and_name_and_type:
|
| 178 |
+
- - dump/raw/tr_no_dev/text
|
| 179 |
+
- text
|
| 180 |
+
- text
|
| 181 |
+
- - duration_info_from_teacher/decode_use_teacher_forcingtrue_train.loss.ave/tr_no_dev/durations
|
| 182 |
+
- durations
|
| 183 |
+
- text_int
|
| 184 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 185 |
+
- speech
|
| 186 |
+
- sound
|
| 187 |
+
train_dtype: float32
|
| 188 |
+
train_shape_file:
|
| 189 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 190 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 191 |
+
tts: fastspeech2
|
| 192 |
+
tts_conf:
|
| 193 |
+
adim: 384
|
| 194 |
+
aheads: 2
|
| 195 |
+
decoder_normalize_before: true
|
| 196 |
+
dlayers: 4
|
| 197 |
+
dunits: 1536
|
| 198 |
+
duration_predictor_chans: 256
|
| 199 |
+
duration_predictor_kernel_size: 3
|
| 200 |
+
duration_predictor_layers: 2
|
| 201 |
+
elayers: 4
|
| 202 |
+
encoder_normalize_before: true
|
| 203 |
+
energy_embed_dropout: 0.0
|
| 204 |
+
energy_embed_kernel_size: 1
|
| 205 |
+
energy_predictor_chans: 256
|
| 206 |
+
energy_predictor_dropout: 0.5
|
| 207 |
+
energy_predictor_kernel_size: 3
|
| 208 |
+
energy_predictor_layers: 2
|
| 209 |
+
eunits: 1536
|
| 210 |
+
init_dec_alpha: 1.0
|
| 211 |
+
init_enc_alpha: 1.0
|
| 212 |
+
init_type: xavier_uniform
|
| 213 |
+
pitch_embed_dropout: 0.0
|
| 214 |
+
pitch_embed_kernel_size: 1
|
| 215 |
+
pitch_predictor_chans: 256
|
| 216 |
+
pitch_predictor_dropout: 0.5
|
| 217 |
+
pitch_predictor_kernel_size: 5
|
| 218 |
+
pitch_predictor_layers: 5
|
| 219 |
+
positionwise_conv_kernel_size: 3
|
| 220 |
+
positionwise_layer_type: conv1d
|
| 221 |
+
postnet_chans: 256
|
| 222 |
+
postnet_filts: 5
|
| 223 |
+
postnet_layers: 5
|
| 224 |
+
reduction_factor: 1
|
| 225 |
+
stop_gradient_from_energy_predictor: false
|
| 226 |
+
stop_gradient_from_pitch_predictor: true
|
| 227 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 228 |
+
transformer_dec_dropout_rate: 0.2
|
| 229 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 230 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 231 |
+
transformer_enc_dropout_rate: 0.2
|
| 232 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 233 |
+
use_masking: true
|
| 234 |
+
use_scaled_pos_enc: true
|
| 235 |
+
unused_parameters: false
|
| 236 |
+
use_amp: false
|
| 237 |
+
use_preprocessor: true
|
| 238 |
+
use_tensorboard: true
|
| 239 |
+
use_wandb: false
|
| 240 |
+
val_scheduler_criterion:
|
| 241 |
+
- valid
|
| 242 |
+
- loss
|
| 243 |
+
valid_batch_bins: null
|
| 244 |
+
valid_batch_size: null
|
| 245 |
+
valid_batch_type: null
|
| 246 |
+
valid_data_path_and_name_and_type:
|
| 247 |
+
- - dump/raw/dev/text
|
| 248 |
+
- text
|
| 249 |
+
- text
|
| 250 |
+
- - duration_info_from_teacher/decode_use_teacher_forcingtrue_train.loss.ave/dev/durations
|
| 251 |
+
- durations
|
| 252 |
+
- text_int
|
| 253 |
+
- - dump/raw/dev/wav.scp
|
| 254 |
+
- speech
|
| 255 |
+
- sound
|
| 256 |
+
valid_max_cache_size: null
|
| 257 |
+
valid_shape_file:
|
| 258 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 259 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 260 |
+
version: 0.10.3a3
|
| 261 |
+
wandb_entity: null
|
| 262 |
+
wandb_id: null
|
| 263 |
+
wandb_model_log_interval: -1
|
| 264 |
+
wandb_name: null
|
| 265 |
+
wandb_project: null
|
| 266 |
+
write_collected_feats: false
|
english/female/model/energy_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3fbd6cac3a0ffa58d825b5107c7272137a33045619b4766222b3dab6ad34290f
|
| 3 |
+
size 770
|
english/female/model/feats_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64edf39613daf282ca50b3738f0667a51f3bdda9feadc3c39cc9bdeb347e5959
|
| 3 |
+
size 1402
|
english/female/model/feats_type
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
raw
|
english/female/model/model.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c637a04f92dab17317a404e6dacf19e4d02c32556a275b5cc2f0eec376392e16
|
| 3 |
+
size 148695034
|
english/female/model/pitch_stats.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:434a06aa750fd55b106aca27d829f654a1822635f42e3d51e8e231ef3f5a4e50
|
| 3 |
+
size 770
|
english/male/model/config.yaml
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
accum_grad: 8
|
| 2 |
+
allow_variable_data_keys: false
|
| 3 |
+
batch_bins: 3000000
|
| 4 |
+
batch_size: 20
|
| 5 |
+
batch_type: numel
|
| 6 |
+
best_model_criterion:
|
| 7 |
+
- - valid
|
| 8 |
+
- loss
|
| 9 |
+
- min
|
| 10 |
+
- - train
|
| 11 |
+
- loss
|
| 12 |
+
- min
|
| 13 |
+
bpemodel: null
|
| 14 |
+
chunk_length: 500
|
| 15 |
+
chunk_shift_ratio: 0.5
|
| 16 |
+
cleaner: null
|
| 17 |
+
collect_stats: false
|
| 18 |
+
config: conf/tuning/train_fastspeech2.yaml
|
| 19 |
+
cudnn_benchmark: false
|
| 20 |
+
cudnn_deterministic: true
|
| 21 |
+
cudnn_enabled: true
|
| 22 |
+
detect_anomaly: false
|
| 23 |
+
dist_backend: nccl
|
| 24 |
+
dist_init_method: env://
|
| 25 |
+
dist_launcher: null
|
| 26 |
+
dist_master_addr: localhost
|
| 27 |
+
dist_master_port: 59485
|
| 28 |
+
dist_rank: 0
|
| 29 |
+
dist_world_size: 8
|
| 30 |
+
distributed: true
|
| 31 |
+
dry_run: false
|
| 32 |
+
early_stopping_criterion:
|
| 33 |
+
- valid
|
| 34 |
+
- loss
|
| 35 |
+
- min
|
| 36 |
+
energy_extract: energy
|
| 37 |
+
energy_extract_conf:
|
| 38 |
+
fs: 22050
|
| 39 |
+
hop_length: 256
|
| 40 |
+
n_fft: 1024
|
| 41 |
+
reduction_factor: 1
|
| 42 |
+
win_length: null
|
| 43 |
+
energy_normalize: global_mvn
|
| 44 |
+
energy_normalize_conf:
|
| 45 |
+
stats_file: /home/speech/Fastspeech2_HS/english/male/model/energy_stats.npz
|
| 46 |
+
feats_extract: fbank
|
| 47 |
+
feats_extract_conf:
|
| 48 |
+
fmax: 8000
|
| 49 |
+
fmin: 0
|
| 50 |
+
fs: 22050
|
| 51 |
+
hop_length: 256
|
| 52 |
+
n_fft: 1024
|
| 53 |
+
n_mels: 80
|
| 54 |
+
win_length: null
|
| 55 |
+
fold_length:
|
| 56 |
+
- 150
|
| 57 |
+
- 204800
|
| 58 |
+
freeze_param: []
|
| 59 |
+
g2p: g2p_en_no_space
|
| 60 |
+
grad_clip: 1.0
|
| 61 |
+
grad_clip_type: 2.0
|
| 62 |
+
grad_noise: false
|
| 63 |
+
ignore_init_mismatch: false
|
| 64 |
+
init_param: []
|
| 65 |
+
iterator_type: sequence
|
| 66 |
+
keep_nbest_models: 5
|
| 67 |
+
local_rank: 0
|
| 68 |
+
log_interval: null
|
| 69 |
+
log_level: INFO
|
| 70 |
+
max_cache_fd: 32
|
| 71 |
+
max_cache_size: 0.0
|
| 72 |
+
max_epoch: 1000
|
| 73 |
+
model_conf: {}
|
| 74 |
+
multiple_iterator: false
|
| 75 |
+
multiprocessing_distributed: true
|
| 76 |
+
ngpu: 1
|
| 77 |
+
no_forward_run: false
|
| 78 |
+
non_linguistic_symbols: null
|
| 79 |
+
normalize: global_mvn
|
| 80 |
+
normalize_conf:
|
| 81 |
+
stats_file: /home/speech/Fastspeech2_HS/english/male/model/feats_stats.npz
|
| 82 |
+
num_att_plot: 3
|
| 83 |
+
num_cache_chunks: 1024
|
| 84 |
+
num_iters_per_epoch: 800
|
| 85 |
+
num_workers: 1
|
| 86 |
+
odim: null
|
| 87 |
+
optim: adam
|
| 88 |
+
optim_conf:
|
| 89 |
+
lr: 1.0
|
| 90 |
+
output_dir: exp/tts_train_fastspeech2_raw_char_None
|
| 91 |
+
patience: null
|
| 92 |
+
pitch_extract: dio
|
| 93 |
+
pitch_extract_conf:
|
| 94 |
+
f0max: 400
|
| 95 |
+
f0min: 40
|
| 96 |
+
fs: 22050
|
| 97 |
+
hop_length: 256
|
| 98 |
+
n_fft: 1024
|
| 99 |
+
reduction_factor: 1
|
| 100 |
+
pitch_normalize: global_mvn
|
| 101 |
+
pitch_normalize_conf:
|
| 102 |
+
stats_file: /home/speech/Fastspeech2_HS/english/male/model/pitch_stats.npz
|
| 103 |
+
pretrain_path: null
|
| 104 |
+
print_config: false
|
| 105 |
+
required:
|
| 106 |
+
- output_dir
|
| 107 |
+
- token_list
|
| 108 |
+
resume: true
|
| 109 |
+
scheduler: noamlr
|
| 110 |
+
scheduler_conf:
|
| 111 |
+
model_size: 384
|
| 112 |
+
warmup_steps: 4000
|
| 113 |
+
seed: 0
|
| 114 |
+
sharded_ddp: false
|
| 115 |
+
sort_batch: descending
|
| 116 |
+
sort_in_batch: descending
|
| 117 |
+
token_list:
|
| 118 |
+
- <blank>
|
| 119 |
+
- <unk>
|
| 120 |
+
- <space>
|
| 121 |
+
- a
|
| 122 |
+
- r
|
| 123 |
+
- n
|
| 124 |
+
- "\u091F"
|
| 125 |
+
- i
|
| 126 |
+
- "\u0921"
|
| 127 |
+
- E
|
| 128 |
+
- s
|
| 129 |
+
- l
|
| 130 |
+
- d
|
| 131 |
+
- w
|
| 132 |
+
- I
|
| 133 |
+
- m
|
| 134 |
+
- k
|
| 135 |
+
- z
|
| 136 |
+
- f
|
| 137 |
+
- "\u0905"
|
| 138 |
+
- h
|
| 139 |
+
- "\u0911"
|
| 140 |
+
- U
|
| 141 |
+
- A
|
| 142 |
+
- .
|
| 143 |
+
- "\u0910"
|
| 144 |
+
- ','
|
| 145 |
+
- p
|
| 146 |
+
- b
|
| 147 |
+
- "\u0919"
|
| 148 |
+
- o
|
| 149 |
+
- g
|
| 150 |
+
- y
|
| 151 |
+
- "\u0936"
|
| 152 |
+
- "\u0914"
|
| 153 |
+
- t
|
| 154 |
+
- u
|
| 155 |
+
- c
|
| 156 |
+
- j
|
| 157 |
+
- '?'
|
| 158 |
+
- '!'
|
| 159 |
+
- q
|
| 160 |
+
- "\u0923"
|
| 161 |
+
- "\u0927"
|
| 162 |
+
- "\u0925"
|
| 163 |
+
- "\u0937"
|
| 164 |
+
- B
|
| 165 |
+
- H
|
| 166 |
+
- P
|
| 167 |
+
- D
|
| 168 |
+
- M
|
| 169 |
+
- v
|
| 170 |
+
- C
|
| 171 |
+
- R
|
| 172 |
+
- "\u0918"
|
| 173 |
+
- "\u0916"
|
| 174 |
+
- <sos/eos>
|
| 175 |
+
token_type: char
|
| 176 |
+
train_data_path_and_name_and_type:
|
| 177 |
+
- - dump/raw/tr_no_dev/text
|
| 178 |
+
- text
|
| 179 |
+
- text
|
| 180 |
+
- - duration_info_from_teacher/decode_use_teacher_forcingtrue_train.loss.ave/tr_no_dev/durations
|
| 181 |
+
- durations
|
| 182 |
+
- text_int
|
| 183 |
+
- - dump/raw/tr_no_dev/wav.scp
|
| 184 |
+
- speech
|
| 185 |
+
- sound
|
| 186 |
+
train_dtype: float32
|
| 187 |
+
train_shape_file:
|
| 188 |
+
- exp/tts_stats_raw_char_None/train/text_shape.char
|
| 189 |
+
- exp/tts_stats_raw_char_None/train/speech_shape
|
| 190 |
+
tts: fastspeech2
|
| 191 |
+
tts_conf:
|
| 192 |
+
adim: 384
|
| 193 |
+
aheads: 2
|
| 194 |
+
decoder_normalize_before: true
|
| 195 |
+
dlayers: 4
|
| 196 |
+
dunits: 1536
|
| 197 |
+
duration_predictor_chans: 256
|
| 198 |
+
duration_predictor_kernel_size: 3
|
| 199 |
+
duration_predictor_layers: 2
|
| 200 |
+
elayers: 4
|
| 201 |
+
encoder_normalize_before: true
|
| 202 |
+
energy_embed_dropout: 0.0
|
| 203 |
+
energy_embed_kernel_size: 1
|
| 204 |
+
energy_predictor_chans: 256
|
| 205 |
+
energy_predictor_dropout: 0.5
|
| 206 |
+
energy_predictor_kernel_size: 3
|
| 207 |
+
energy_predictor_layers: 2
|
| 208 |
+
eunits: 1536
|
| 209 |
+
init_dec_alpha: 1.0
|
| 210 |
+
init_enc_alpha: 1.0
|
| 211 |
+
init_type: xavier_uniform
|
| 212 |
+
pitch_embed_dropout: 0.0
|
| 213 |
+
pitch_embed_kernel_size: 1
|
| 214 |
+
pitch_predictor_chans: 256
|
| 215 |
+
pitch_predictor_dropout: 0.5
|
| 216 |
+
pitch_predictor_kernel_size: 5
|
| 217 |
+
pitch_predictor_layers: 5
|
| 218 |
+
positionwise_conv_kernel_size: 3
|
| 219 |
+
positionwise_layer_type: conv1d
|
| 220 |
+
postnet_chans: 256
|
| 221 |
+
postnet_filts: 5
|
| 222 |
+
postnet_layers: 5
|
| 223 |
+
reduction_factor: 1
|
| 224 |
+
stop_gradient_from_energy_predictor: false
|
| 225 |
+
stop_gradient_from_pitch_predictor: true
|
| 226 |
+
transformer_dec_attn_dropout_rate: 0.2
|
| 227 |
+
transformer_dec_dropout_rate: 0.2
|
| 228 |
+
transformer_dec_positional_dropout_rate: 0.2
|
| 229 |
+
transformer_enc_attn_dropout_rate: 0.2
|
| 230 |
+
transformer_enc_dropout_rate: 0.2
|
| 231 |
+
transformer_enc_positional_dropout_rate: 0.2
|
| 232 |
+
use_masking: true
|
| 233 |
+
use_scaled_pos_enc: true
|
| 234 |
+
unused_parameters: false
|
| 235 |
+
use_amp: false
|
| 236 |
+
use_preprocessor: true
|
| 237 |
+
use_tensorboard: true
|
| 238 |
+
use_wandb: false
|
| 239 |
+
val_scheduler_criterion:
|
| 240 |
+
- valid
|
| 241 |
+
- loss
|
| 242 |
+
valid_batch_bins: null
|
| 243 |
+
valid_batch_size: null
|
| 244 |
+
valid_batch_type: null
|
| 245 |
+
valid_data_path_and_name_and_type:
|
| 246 |
+
- - dump/raw/dev/text
|
| 247 |
+
- text
|
| 248 |
+
- text
|
| 249 |
+
- - duration_info_from_teacher/decode_use_teacher_forcingtrue_train.loss.ave/dev/durations
|
| 250 |
+
- durations
|
| 251 |
+
- text_int
|
| 252 |
+
- - dump/raw/dev/wav.scp
|
| 253 |
+
- speech
|
| 254 |
+
- sound
|
| 255 |
+
valid_max_cache_size: null
|
| 256 |
+
valid_shape_file:
|
| 257 |
+
- exp/tts_stats_raw_char_None/valid/text_shape.char
|
| 258 |
+
- exp/tts_stats_raw_char_None/valid/speech_shape
|
| 259 |
+
version: 0.10.3a3
|
| 260 |
+
wandb_entity: null
|
| 261 |
+
wandb_id: null
|
| 262 |
+
wandb_model_log_interval: -1
|
| 263 |
+
wandb_name: null
|
| 264 |
+
wandb_project: null
|
| 265 |
+
write_collected_feats: false
|