Hugging Face's logo

RESPIN
/
Telugu_LanguageModels_for_ASR

Model card Files
xet
Community
viks66 commited on
Commit
80731f5
·
1 Parent(s): 59b4e0d

add configs

Browse files
Files changed (3) hide show
  1. espnet_models/exp/lm_combined_a_m/config.yaml +187 -0
  2. espnet_models/exp/lm_combined_automatic/config.yaml +187 -0
  3. espnet_models/exp/lm_combined_manual/config.yaml +189 -0
espnet_models/exp/lm_combined_a_m/config.yaml ADDED
@@ -0,0 +1,187 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ config: conf/lm_config.yaml
2
+ print_config: false
3
+ log_level: INFO
4
+ dry_run: false
5
+ iterator_type: sequence
6
+ output_dir: exp/lm_combined_a_m
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: null
14
+ dist_rank: null
15
+ local_rank: 0
16
+ dist_master_addr: null
17
+ dist_master_port: null
18
+ dist_launcher: null
19
+ multiprocessing_distributed: false
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: 25
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
+ keep_nbest_models: 10
41
+ grad_clip: 1.0
42
+ grad_clip_type: 2.0
43
+ grad_noise: false
44
+ accum_grad: 2
45
+ no_forward_run: false
46
+ resume: true
47
+ train_dtype: float32
48
+ use_amp: true
49
+ log_interval: null
50
+ use_tensorboard: true
51
+ use_wandb: false
52
+ wandb_project: null
53
+ wandb_id: null
54
+ wandb_entity: null
55
+ wandb_name: null
56
+ wandb_model_log_interval: -1
57
+ detect_anomaly: false
58
+ pretrain_path: null
59
+ init_param: []
60
+ ignore_init_mismatch: false
61
+ freeze_param: []
62
+ num_iters_per_epoch: null
63
+ batch_size: 32
64
+ valid_batch_size: null
65
+ batch_bins: 1000000
66
+ valid_batch_bins: null
67
+ train_shape_file:
68
+ - exp/lm_stats_te_char/train/text_shape.char
69
+ valid_shape_file:
70
+ - exp/lm_stats_te_char/valid/text_shape.char
71
+ batch_type: sorted
72
+ valid_batch_type: null
73
+ fold_length:
74
+ - 150
75
+ sort_in_batch: descending
76
+ sort_batch: descending
77
+ multiple_iterator: false
78
+ chunk_length: 500
79
+ chunk_shift_ratio: 0.5
80
+ num_cache_chunks: 1024
81
+ train_data_path_and_name_and_type:
82
+ - - dump/raw/lm_train.txt
83
+ - text
84
+ - text
85
+ valid_data_path_and_name_and_type:
86
+ - - dump/raw/dev/text
87
+ - text
88
+ - text
89
+ allow_variable_data_keys: false
90
+ max_cache_size: 0.0
91
+ max_cache_fd: 32
92
+ valid_max_cache_size: null
93
+ optim: adam
94
+ optim_conf:
95
+ lr: 0.005
96
+ scheduler: warmuplr
97
+ scheduler_conf:
98
+ warmup_steps: 25000
99
+ token_list:
100
+ - <blank>
101
+ - <unk>
102
+ - <space>
103
+ - ్
104
+ - ి
105
+ - ా
106
+ - ు
107
+ - ర
108
+ - న
109
+ - ం
110
+ - ల
111
+ - క
112
+ - త
113
+ - ప
114
+ - వ
115
+ - స
116
+ - య
117
+ - ద
118
+ - మ
119
+ - చ
120
+ - ట
121
+ - ే
122
+ - ో
123
+ - గ
124
+ - డ
125
+ - ీ
126
+ - ె
127
+ - అ
128
+ - బ
129
+ - ధ
130
+ - జ
131
+ - ొ
132
+ - ూ
133
+ - ఉ
134
+ - ై
135
+ - ష
136
+ - ణ
137
+ - శ
138
+ - భ
139
+ - ఆ
140
+ - హ
141
+ - ఇ
142
+ - థ
143
+ - ఎ
144
+ - ఫ
145
+ - ఈ
146
+ - ఒ
147
+ - ళ
148
+ - ఖ
149
+ - ృ
150
+ - ఏ
151
+ - ౌ
152
+ - ౖ
153
+ - ఘ
154
+ - ఐ
155
+ - ఛ
156
+ - ఓ
157
+ - ఠ
158
+ - ఞ
159
+ - ఊ
160
+ - ఔ
161
+ - ఢ
162
+ - ఋ
163
+ - ః
164
+ - <sos/eos>
165
+ init: null
166
+ model_conf:
167
+ ignore_id: 0
168
+ use_preprocessor: true
169
+ token_type: char
170
+ bpemodel: null
171
+ non_linguistic_symbols: null
172
+ cleaner: null
173
+ g2p: null
174
+ lm: transformer
175
+ lm_conf:
176
+ pos_enc: null
177
+ embed_unit: 128
178
+ att_unit: 512
179
+ head: 4
180
+ unit: 1024
181
+ layer: 8
182
+ dropout_rate: 0.0
183
+ required:
184
+ - output_dir
185
+ - token_list
186
+ version: 0.10.2a1
187
+ distributed: false
espnet_models/exp/lm_combined_automatic/config.yaml ADDED
@@ -0,0 +1,187 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ config: conf/lm_config.yaml
2
+ print_config: false
3
+ log_level: INFO
4
+ dry_run: false
5
+ iterator_type: sequence
6
+ output_dir: exp/lm_combined_automatic
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: null
14
+ dist_rank: null
15
+ local_rank: 0
16
+ dist_master_addr: null
17
+ dist_master_port: null
18
+ dist_launcher: null
19
+ multiprocessing_distributed: false
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: 25
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
+ keep_nbest_models: 10
41
+ grad_clip: 1.0
42
+ grad_clip_type: 2.0
43
+ grad_noise: false
44
+ accum_grad: 2
45
+ no_forward_run: false
46
+ resume: true
47
+ train_dtype: float32
48
+ use_amp: true
49
+ log_interval: null
50
+ use_tensorboard: true
51
+ use_wandb: false
52
+ wandb_project: null
53
+ wandb_id: null
54
+ wandb_entity: null
55
+ wandb_name: null
56
+ wandb_model_log_interval: -1
57
+ detect_anomaly: false
58
+ pretrain_path: null
59
+ init_param: []
60
+ ignore_init_mismatch: false
61
+ freeze_param: []
62
+ num_iters_per_epoch: null
63
+ batch_size: 32
64
+ valid_batch_size: null
65
+ batch_bins: 1000000
66
+ valid_batch_bins: null
67
+ train_shape_file:
68
+ - exp/lm_stats_te_char/train/text_shape.char
69
+ valid_shape_file:
70
+ - exp/lm_stats_te_char/valid/text_shape.char
71
+ batch_type: sorted
72
+ valid_batch_type: null
73
+ fold_length:
74
+ - 150
75
+ sort_in_batch: descending
76
+ sort_batch: descending
77
+ multiple_iterator: false
78
+ chunk_length: 500
79
+ chunk_shift_ratio: 0.5
80
+ num_cache_chunks: 1024
81
+ train_data_path_and_name_and_type:
82
+ - - dump/raw/lm_train.txt
83
+ - text
84
+ - text
85
+ valid_data_path_and_name_and_type:
86
+ - - dump/raw/dev/text
87
+ - text
88
+ - text
89
+ allow_variable_data_keys: false
90
+ max_cache_size: 0.0
91
+ max_cache_fd: 32
92
+ valid_max_cache_size: null
93
+ optim: adam
94
+ optim_conf:
95
+ lr: 0.005
96
+ scheduler: warmuplr
97
+ scheduler_conf:
98
+ warmup_steps: 25000
99
+ token_list:
100
+ - <blank>
101
+ - <unk>
102
+ - <space>
103
+ - ్
104
+ - ి
105
+ - ా
106
+ - ు
107
+ - ర
108
+ - న
109
+ - ం
110
+ - ల
111
+ - క
112
+ - త
113
+ - ప
114
+ - వ
115
+ - స
116
+ - య
117
+ - ద
118
+ - మ
119
+ - చ
120
+ - ట
121
+ - ో
122
+ - ే
123
+ - గ
124
+ - డ
125
+ - ీ
126
+ - ె
127
+ - అ
128
+ - బ
129
+ - ధ
130
+ - జ
131
+ - ూ
132
+ - ొ
133
+ - ఉ
134
+ - ై
135
+ - ష
136
+ - శ
137
+ - ణ
138
+ - భ
139
+ - ఆ
140
+ - హ
141
+ - ఇ
142
+ - థ
143
+ - ఎ
144
+ - ఫ
145
+ - ఈ
146
+ - ఒ
147
+ - ళ
148
+ - ఖ
149
+ - ృ
150
+ - ఏ
151
+ - ౌ
152
+ - ౖ
153
+ - ఐ
154
+ - ఘ
155
+ - ఛ
156
+ - ఓ
157
+ - ఠ
158
+ - ఞ
159
+ - ఊ
160
+ - ఔ
161
+ - ఢ
162
+ - ః
163
+ - ఋ
164
+ - <sos/eos>
165
+ init: null
166
+ model_conf:
167
+ ignore_id: 0
168
+ use_preprocessor: true
169
+ token_type: char
170
+ bpemodel: null
171
+ non_linguistic_symbols: null
172
+ cleaner: null
173
+ g2p: null
174
+ lm: transformer
175
+ lm_conf:
176
+ pos_enc: null
177
+ embed_unit: 128
178
+ att_unit: 512
179
+ head: 4
180
+ unit: 1024
181
+ layer: 8
182
+ dropout_rate: 0.0
183
+ required:
184
+ - output_dir
185
+ - token_list
186
+ version: 0.10.2a1
187
+ distributed: false
espnet_models/exp/lm_combined_manual/config.yaml ADDED
@@ -0,0 +1,189 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ config: conf/lm_config.yaml
2
+ print_config: false
3
+ log_level: INFO
4
+ dry_run: false
5
+ iterator_type: sequence
6
+ output_dir: exp/lm_combined_manual
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: null
14
+ dist_rank: null
15
+ local_rank: 0
16
+ dist_master_addr: null
17
+ dist_master_port: null
18
+ dist_launcher: null
19
+ multiprocessing_distributed: false
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: 25
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
+ keep_nbest_models: 10
41
+ nbest_averaging_interval: 0
42
+ grad_clip: 1.0
43
+ grad_clip_type: 2.0
44
+ grad_noise: false
45
+ accum_grad: 2
46
+ no_forward_run: false
47
+ resume: true
48
+ train_dtype: float32
49
+ use_amp: true
50
+ log_interval: null
51
+ use_matplotlib: true
52
+ use_tensorboard: true
53
+ use_wandb: false
54
+ wandb_project: null
55
+ wandb_id: null
56
+ wandb_entity: null
57
+ wandb_name: null
58
+ wandb_model_log_interval: -1
59
+ detect_anomaly: false
60
+ pretrain_path: null
61
+ init_param: []
62
+ ignore_init_mismatch: false
63
+ freeze_param: []
64
+ num_iters_per_epoch: null
65
+ batch_size: 32
66
+ valid_batch_size: null
67
+ batch_bins: 1000000
68
+ valid_batch_bins: null
69
+ train_shape_file:
70
+ - exp/lm_stats_te_char/train/text_shape.char
71
+ valid_shape_file:
72
+ - exp/lm_stats_te_char/valid/text_shape.char
73
+ batch_type: sorted
74
+ valid_batch_type: null
75
+ fold_length:
76
+ - 150
77
+ sort_in_batch: descending
78
+ sort_batch: descending
79
+ multiple_iterator: false
80
+ chunk_length: 500
81
+ chunk_shift_ratio: 0.5
82
+ num_cache_chunks: 1024
83
+ train_data_path_and_name_and_type:
84
+ - - dump/raw/lm_train.txt
85
+ - text
86
+ - text
87
+ valid_data_path_and_name_and_type:
88
+ - - dump/raw/dev/text
89
+ - text
90
+ - text
91
+ allow_variable_data_keys: false
92
+ max_cache_size: 0.0
93
+ max_cache_fd: 32
94
+ valid_max_cache_size: null
95
+ optim: adam
96
+ optim_conf:
97
+ lr: 0.005
98
+ scheduler: warmuplr
99
+ scheduler_conf:
100
+ warmup_steps: 25000
101
+ token_list:
102
+ - <blank>
103
+ - <unk>
104
+ - <space>
105
+ - ్
106
+ - ి
107
+ - ా
108
+ - ు
109
+ - ర
110
+ - న
111
+ - ం
112
+ - ల
113
+ - క
114
+ - త
115
+ - ప
116
+ - వ
117
+ - స
118
+ - య
119
+ - ద
120
+ - మ
121
+ - చ
122
+ - ట
123
+ - గ
124
+ - ే
125
+ - ో
126
+ - డ
127
+ - ీ
128
+ - ె
129
+ - అ
130
+ - బ
131
+ - ధ
132
+ - జ
133
+ - ొ
134
+ - ఉ
135
+ - ూ
136
+ - ై
137
+ - ణ
138
+ - ష
139
+ - శ
140
+ - భ
141
+ - ఆ
142
+ - హ
143
+ - ఇ
144
+ - థ
145
+ - ఎ
146
+ - ఫ
147
+ - ఈ
148
+ - ఒ
149
+ - ళ
150
+ - ఖ
151
+ - ృ
152
+ - ఏ
153
+ - ౌ
154
+ - ఘ
155
+ - ౖ
156
+ - ఐ
157
+ - ఛ
158
+ - ఓ
159
+ - ఠ
160
+ - ఞ
161
+ - ఊ
162
+ - ఔ
163
+ - ఢ
164
+ - ఋ
165
+ - ః
166
+ - <sos/eos>
167
+ init: null
168
+ model_conf:
169
+ ignore_id: 0
170
+ use_preprocessor: true
171
+ token_type: char
172
+ bpemodel: null
173
+ non_linguistic_symbols: null
174
+ cleaner: null
175
+ g2p: null
176
+ lm: transformer
177
+ lm_conf:
178
+ pos_enc: null
179
+ embed_unit: 128
180
+ att_unit: 512
181
+ head: 4
182
+ unit: 1024
183
+ layer: 8
184
+ dropout_rate: 0.0
185
+ required:
186
+ - output_dir
187
+ - token_list
188
+ version: '202204'
189
+ distributed: false