File size: 4,405 Bytes
febac81 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 |
config: conf/tuning/train_gst_fastspeech.yaml
print_config: false
log_level: INFO
dry_run: false
iterator_type: sequence
output_dir: exp/tts_train_gst_fastspeech_raw_phn_tacotron_g2p_en_no_space
ngpu: 1
seed: 0
num_workers: 1
num_att_plot: 3
dist_backend: nccl
dist_init_method: env://
dist_world_size: null
dist_rank: null
local_rank: 0
dist_master_addr: null
dist_master_port: null
dist_launcher: null
multiprocessing_distributed: false
cudnn_enabled: true
cudnn_benchmark: false
cudnn_deterministic: true
collect_stats: false
write_collected_feats: false
max_epoch: 1000
patience: null
val_scheduler_criterion:
- valid
- loss
early_stopping_criterion:
- valid
- loss
- min
best_model_criterion:
- - valid
- loss
- min
- - train
- loss
- min
keep_nbest_models: 5
grad_clip: 1.0
grad_noise: false
accum_grad: 2
no_forward_run: false
resume: true
train_dtype: float32
log_interval: null
pretrain_path: []
pretrain_key: []
num_iters_per_epoch: null
batch_size: 20
valid_batch_size: null
batch_bins: 2400000
valid_batch_bins: null
train_shape_file:
- exp/tts_stats_raw_phn_tacotron_g2p_en_no_space/train/text_shape.phn
- exp/tts_train_gst_tacotron2_raw_phn_tacotron_g2p_en_no_space/decode_use_att_constrainttrue_train.loss.best/tr_no_dev/speech_shape
valid_shape_file:
- exp/tts_stats_raw_phn_tacotron_g2p_en_no_space/valid/text_shape.phn
- exp/tts_train_gst_tacotron2_raw_phn_tacotron_g2p_en_no_space/decode_use_att_constrainttrue_train.loss.best/dev/speech_shape
batch_type: numel
valid_batch_type: null
fold_length:
- 150
- 800
sort_in_batch: descending
sort_batch: descending
multiple_iterator: false
chunk_length: 500
chunk_shift_ratio: 0.5
num_cache_chunks: 1024
train_data_path_and_name_and_type:
- - dump/raw/tr_no_dev/text
- text
- text
- - exp/tts_train_gst_tacotron2_raw_phn_tacotron_g2p_en_no_space/decode_use_att_constrainttrue_train.loss.best/tr_no_dev/denorm/feats.scp
- speech
- npy
- - exp/tts_train_gst_tacotron2_raw_phn_tacotron_g2p_en_no_space/decode_use_att_constrainttrue_train.loss.best/tr_no_dev/durations
- durations
- text_int
valid_data_path_and_name_and_type:
- - dump/raw/dev/text
- text
- text
- - exp/tts_train_gst_tacotron2_raw_phn_tacotron_g2p_en_no_space/decode_use_att_constrainttrue_train.loss.best/dev/denorm/feats.scp
- speech
- npy
- - exp/tts_train_gst_tacotron2_raw_phn_tacotron_g2p_en_no_space/decode_use_att_constrainttrue_train.loss.best/dev/durations
- durations
- text_int
allow_variable_data_keys: false
max_cache_size: 0.0
valid_max_cache_size: null
optim: adam
optim_conf:
lr: 1.0
scheduler: noamlr
scheduler_conf:
model_size: 384
warmup_steps: 4000
token_list:
- <blank>
- <unk>
- OY0
- ''''
- OY2
- ER2
- UH0
- '!'
- EY0
- AW0
- AA0
- UH2
- UW2
- AY0
- AO2
- AO0
- AE2
- AH2
- AE0
- AA2
- IY2
- EH0
- AW2
- ZH
- AY2
- OY1
- IH2
- UW0
- EY2
- EH2
- OW2
- OW0
- '?'
- CH
- ER1
- TH
- UH1
- AW1
- JH
- Y
- SH
- NG
- ','
- G
- OW1
- AO1
- IY0
- UW1
- EY1
- AY1
- HH
- F
- ER0
- V
- P
- B
- AH1
- IY1
- IH0
- AA1
- EH1
- AE1
- M
- W
- K
- DH
- Z
- .
- L
- D
- IH1
- R
- S
- N
- T
- AH0
- <sos/eos>
odim: 80
model_conf: {}
use_preprocessor: true
token_type: phn
bpemodel: null
non_linguistic_symbols: null
cleaner: tacotron
g2p: g2p_en_no_space
feats_extract: null
feats_extract_conf: null
normalize: global_mvn
normalize_conf:
stats_file: exp/tts_stats_raw_phn_tacotron_g2p_en_no_space/train/feats_stats.npz
tts: fastspeech
tts_conf:
adim: 384
aheads: 4
elayers: 6
eunits: 1536
dlayers: 6
dunits: 1536
positionwise_layer_type: conv1d
positionwise_conv_kernel_size: 3
duration_predictor_layers: 2
duration_predictor_chans: 384
duration_predictor_kernel_size: 3
postnet_layers: 5
postnet_filts: 5
postnet_chans: 256
use_masking: true
use_scaled_pos_enc: true
encoder_normalize_before: false
decoder_normalize_before: false
reduction_factor: 1
init_type: xavier_uniform
init_enc_alpha: 1.0
init_dec_alpha: 1.0
transformer_enc_dropout_rate: 0.1
transformer_enc_positional_dropout_rate: 0.1
transformer_enc_attn_dropout_rate: 0.1
transformer_dec_dropout_rate: 0.1
transformer_dec_positional_dropout_rate: 0.1
transformer_dec_attn_dropout_rate: 0.1
use_gst: true
gst_heads: 8
gst_tokens: 128
required:
- output_dir
- token_list
distributed: false
|