Hugging Face's logo

codeceejay
/
HIYACCENT_Wav2Vec2

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Inference Endpoints
Model card Files Metrics Community
codeceejay commited on
Commit
6bf29e3
•
1 Parent(s): e32aa20

add tokenizer

Browse files
Files changed (22) hide show
  1. checkpoint-3500/config.json +88 -0
  2. checkpoint-3500/optimizer.pt +3 -0
  3. checkpoint-3500/preprocessor_config.json +9 -0
  4. checkpoint-3500/pytorch_model.bin +3 -0
  5. checkpoint-3500/rng_state.pth +3 -0
  6. checkpoint-3500/scheduler.pt +3 -0
  7. checkpoint-3500/trainer_state.json +121 -0
  8. checkpoint-3500/training_args.bin +3 -0
  9. checkpoint-4000/config.json +88 -0
  10. checkpoint-4000/optimizer.pt +3 -0
  11. checkpoint-4000/preprocessor_config.json +9 -0
  12. checkpoint-4000/pytorch_model.bin +3 -0
  13. checkpoint-4000/rng_state.pth +3 -0
  14. checkpoint-4000/scheduler.pt +3 -0
  15. checkpoint-4000/trainer_state.json +136 -0
  16. checkpoint-4000/training_args.bin +3 -0
  17. config.json +88 -0
  18. preprocessor_config.json +9 -0
  19. pytorch_model.bin +3 -0
  20. runs/Feb17_10-01-55_5fabccd8880d/1645092213.5924044/events.out.tfevents.1645092213.5fabccd8880d.84.1 +3 -0
  21. runs/Feb17_10-01-55_5fabccd8880d/events.out.tfevents.1645092213.5fabccd8880d.84.0 +3 -0
  22. training_args.bin +3 -0
checkpoint-3500/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base",
3
+ "activation_dropout": 0.0,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2ForCTC"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "codevector_dim": 256,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": false,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "mean",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": false,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_norm": "group",
48
+ "feat_proj_dropout": 0.1,
49
+ "feat_quantizer_dropout": 0.0,
50
+ "final_dropout": 0.0,
51
+ "freeze_feat_extract_train": true,
52
+ "hidden_act": "gelu",
53
+ "hidden_dropout": 0.1,
54
+ "hidden_size": 768,
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "layer_norm_eps": 1e-05,
58
+ "layerdrop": 0.0,
59
+ "mask_channel_length": 10,
60
+ "mask_channel_min_space": 1,
61
+ "mask_channel_other": 0.0,
62
+ "mask_channel_prob": 0.0,
63
+ "mask_channel_selection": "static",
64
+ "mask_feature_length": 10,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_space": 1,
68
+ "mask_time_other": 0.0,
69
+ "mask_time_prob": 0.05,
70
+ "mask_time_selection": "static",
71
+ "model_type": "wav2vec2",
72
+ "no_mask_channel_overlap": false,
73
+ "no_mask_time_overlap": false,
74
+ "num_attention_heads": 12,
75
+ "num_codevector_groups": 2,
76
+ "num_codevectors_per_group": 320,
77
+ "num_conv_pos_embedding_groups": 16,
78
+ "num_conv_pos_embeddings": 128,
79
+ "num_feat_extract_layers": 7,
80
+ "num_hidden_layers": 12,
81
+ "num_negatives": 100,
82
+ "pad_token_id": 29,
83
+ "proj_codevector_dim": 256,
84
+ "torch_dtype": "float32",
85
+ "transformers_version": "4.11.3",
86
+ "use_weighted_layer_sum": false,
87
+ "vocab_size": 32
88
+ }
checkpoint-3500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99861ace4630fda396ffdc081c88874bb2ca77dff2253f16d411f12c58023d0a
3
+ size 721685265
checkpoint-3500/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-3500/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3567526ae7fc47ee5f40c0d051b76a6e71716f099939c4d35ac36e58e5d10d02
3
+ size 377670039
checkpoint-3500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e6d8e25fb9812a2b3025baf4bc731519fdeffc9eb86697297e4247c02ba96ed
3
+ size 14567
checkpoint-3500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5475e293dbb6816c0bdbf454ef51c5f0d5aca3d2a9249551e4db5b35cf878315
3
+ size 623
checkpoint-3500/trainer_state.json ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 24.137931034482758,
5
+ "global_step": 3500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 3.45,
12
+ "learning_rate": 5e-05,
13
+ "loss": 3.4838,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.45,
18
+ "eval_loss": 1.1867836713790894,
19
+ "eval_runtime": 80.0488,
20
+ "eval_samples_per_second": 20.987,
21
+ "eval_steps_per_second": 2.623,
22
+ "eval_wer": 0.8338501826200814,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 6.9,
27
+ "learning_rate": 0.0001,
28
+ "loss": 0.5623,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 6.9,
33
+ "eval_loss": 0.4056074321269989,
34
+ "eval_runtime": 78.6534,
35
+ "eval_samples_per_second": 21.36,
36
+ "eval_steps_per_second": 2.67,
37
+ "eval_wer": 0.3664116876852043,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 10.34,
42
+ "learning_rate": 8.507462686567164e-05,
43
+ "loss": 0.2018,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 10.34,
48
+ "eval_loss": 0.407957524061203,
49
+ "eval_runtime": 78.7154,
50
+ "eval_samples_per_second": 21.343,
51
+ "eval_steps_per_second": 2.668,
52
+ "eval_wer": 0.3300254978981462,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 13.79,
57
+ "learning_rate": 7.014925373134329e-05,
58
+ "loss": 0.1217,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 13.79,
63
+ "eval_loss": 0.38568004965782166,
64
+ "eval_runtime": 78.9883,
65
+ "eval_samples_per_second": 21.269,
66
+ "eval_steps_per_second": 2.659,
67
+ "eval_wer": 0.3065949968989043,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 17.24,
72
+ "learning_rate": 5.5223880597014934e-05,
73
+ "loss": 0.0829,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 17.24,
78
+ "eval_loss": 0.42486006021499634,
79
+ "eval_runtime": 79.0665,
80
+ "eval_samples_per_second": 21.248,
81
+ "eval_steps_per_second": 2.656,
82
+ "eval_wer": 0.2981186685962373,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 20.69,
87
+ "learning_rate": 4.029850746268657e-05,
88
+ "loss": 0.0626,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 20.69,
93
+ "eval_loss": 0.4198434352874756,
94
+ "eval_runtime": 78.3866,
95
+ "eval_samples_per_second": 21.432,
96
+ "eval_steps_per_second": 2.679,
97
+ "eval_wer": 0.2981186685962373,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 24.14,
102
+ "learning_rate": 2.537313432835821e-05,
103
+ "loss": 0.0488,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 24.14,
108
+ "eval_loss": 0.4277154803276062,
109
+ "eval_runtime": 78.4897,
110
+ "eval_samples_per_second": 21.404,
111
+ "eval_steps_per_second": 2.676,
112
+ "eval_wer": 0.29115843153469784,
113
+ "step": 3500
114
+ }
115
+ ],
116
+ "max_steps": 4350,
117
+ "num_train_epochs": 30,
118
+ "total_flos": 3.2107549243060224e+18,
119
+ "trial_name": null,
120
+ "trial_params": null
121
+ }
checkpoint-3500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1b07d1dd4cc6493def1bd7e8913100e68f098e1ae5794305fd1ea27af2e811b
3
+ size 2799
checkpoint-4000/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base",
3
+ "activation_dropout": 0.0,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2ForCTC"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "codevector_dim": 256,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": false,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "mean",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": false,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_norm": "group",
48
+ "feat_proj_dropout": 0.1,
49
+ "feat_quantizer_dropout": 0.0,
50
+ "final_dropout": 0.0,
51
+ "freeze_feat_extract_train": true,
52
+ "hidden_act": "gelu",
53
+ "hidden_dropout": 0.1,
54
+ "hidden_size": 768,
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "layer_norm_eps": 1e-05,
58
+ "layerdrop": 0.0,
59
+ "mask_channel_length": 10,
60
+ "mask_channel_min_space": 1,
61
+ "mask_channel_other": 0.0,
62
+ "mask_channel_prob": 0.0,
63
+ "mask_channel_selection": "static",
64
+ "mask_feature_length": 10,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_space": 1,
68
+ "mask_time_other": 0.0,
69
+ "mask_time_prob": 0.05,
70
+ "mask_time_selection": "static",
71
+ "model_type": "wav2vec2",
72
+ "no_mask_channel_overlap": false,
73
+ "no_mask_time_overlap": false,
74
+ "num_attention_heads": 12,
75
+ "num_codevector_groups": 2,
76
+ "num_codevectors_per_group": 320,
77
+ "num_conv_pos_embedding_groups": 16,
78
+ "num_conv_pos_embeddings": 128,
79
+ "num_feat_extract_layers": 7,
80
+ "num_hidden_layers": 12,
81
+ "num_negatives": 100,
82
+ "pad_token_id": 29,
83
+ "proj_codevector_dim": 256,
84
+ "torch_dtype": "float32",
85
+ "transformers_version": "4.11.3",
86
+ "use_weighted_layer_sum": false,
87
+ "vocab_size": 32
88
+ }
checkpoint-4000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a0aa53290168acb9ffb5355e781ab6f9647e17c831919fc2b413036cf1845be
3
+ size 721685265
checkpoint-4000/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-4000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5502a20acaf9d32bf3da8370c270269794a492a40863afe8f6b7caadc8a58e6
3
+ size 377670039
checkpoint-4000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:581af1c9cd357cddd65c37fd2e4420386c72edc321fab709ee4cd6f44d744fe9
3
+ size 14567
checkpoint-4000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40568bd6dae503a5711cf588bbce02ac3baf9f55fe60ab8b769e47c7da963ead
3
+ size 623
checkpoint-4000/trainer_state.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 27.586206896551722,
5
+ "global_step": 4000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 3.45,
12
+ "learning_rate": 5e-05,
13
+ "loss": 3.4838,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.45,
18
+ "eval_loss": 1.1867836713790894,
19
+ "eval_runtime": 80.0488,
20
+ "eval_samples_per_second": 20.987,
21
+ "eval_steps_per_second": 2.623,
22
+ "eval_wer": 0.8338501826200814,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 6.9,
27
+ "learning_rate": 0.0001,
28
+ "loss": 0.5623,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 6.9,
33
+ "eval_loss": 0.4056074321269989,
34
+ "eval_runtime": 78.6534,
35
+ "eval_samples_per_second": 21.36,
36
+ "eval_steps_per_second": 2.67,
37
+ "eval_wer": 0.3664116876852043,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 10.34,
42
+ "learning_rate": 8.507462686567164e-05,
43
+ "loss": 0.2018,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 10.34,
48
+ "eval_loss": 0.407957524061203,
49
+ "eval_runtime": 78.7154,
50
+ "eval_samples_per_second": 21.343,
51
+ "eval_steps_per_second": 2.668,
52
+ "eval_wer": 0.3300254978981462,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 13.79,
57
+ "learning_rate": 7.014925373134329e-05,
58
+ "loss": 0.1217,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 13.79,
63
+ "eval_loss": 0.38568004965782166,
64
+ "eval_runtime": 78.9883,
65
+ "eval_samples_per_second": 21.269,
66
+ "eval_steps_per_second": 2.659,
67
+ "eval_wer": 0.3065949968989043,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 17.24,
72
+ "learning_rate": 5.5223880597014934e-05,
73
+ "loss": 0.0829,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 17.24,
78
+ "eval_loss": 0.42486006021499634,
79
+ "eval_runtime": 79.0665,
80
+ "eval_samples_per_second": 21.248,
81
+ "eval_steps_per_second": 2.656,
82
+ "eval_wer": 0.2981186685962373,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 20.69,
87
+ "learning_rate": 4.029850746268657e-05,
88
+ "loss": 0.0626,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 20.69,
93
+ "eval_loss": 0.4198434352874756,
94
+ "eval_runtime": 78.3866,
95
+ "eval_samples_per_second": 21.432,
96
+ "eval_steps_per_second": 2.679,
97
+ "eval_wer": 0.2981186685962373,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 24.14,
102
+ "learning_rate": 2.537313432835821e-05,
103
+ "loss": 0.0488,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 24.14,
108
+ "eval_loss": 0.4277154803276062,
109
+ "eval_runtime": 78.4897,
110
+ "eval_samples_per_second": 21.404,
111
+ "eval_steps_per_second": 2.676,
112
+ "eval_wer": 0.29115843153469784,
113
+ "step": 3500
114
+ },
115
+ {
116
+ "epoch": 27.59,
117
+ "learning_rate": 1.0447761194029851e-05,
118
+ "loss": 0.0393,
119
+ "step": 4000
120
+ },
121
+ {
122
+ "epoch": 27.59,
123
+ "eval_loss": 0.4434719979763031,
124
+ "eval_runtime": 78.6733,
125
+ "eval_samples_per_second": 21.354,
126
+ "eval_steps_per_second": 2.669,
127
+ "eval_wer": 0.2872992901936462,
128
+ "step": 4000
129
+ }
130
+ ],
131
+ "max_steps": 4350,
132
+ "num_train_epochs": 30,
133
+ "total_flos": 3.6694645512377933e+18,
134
+ "trial_name": null,
135
+ "trial_params": null
136
+ }
checkpoint-4000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1b07d1dd4cc6493def1bd7e8913100e68f098e1ae5794305fd1ea27af2e811b
3
+ size 2799
config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base",
3
+ "activation_dropout": 0.0,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2ForCTC"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "codevector_dim": 256,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": false,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "mean",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": false,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_norm": "group",
48
+ "feat_proj_dropout": 0.1,
49
+ "feat_quantizer_dropout": 0.0,
50
+ "final_dropout": 0.0,
51
+ "freeze_feat_extract_train": true,
52
+ "hidden_act": "gelu",
53
+ "hidden_dropout": 0.1,
54
+ "hidden_size": 768,
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "layer_norm_eps": 1e-05,
58
+ "layerdrop": 0.0,
59
+ "mask_channel_length": 10,
60
+ "mask_channel_min_space": 1,
61
+ "mask_channel_other": 0.0,
62
+ "mask_channel_prob": 0.0,
63
+ "mask_channel_selection": "static",
64
+ "mask_feature_length": 10,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_space": 1,
68
+ "mask_time_other": 0.0,
69
+ "mask_time_prob": 0.05,
70
+ "mask_time_selection": "static",
71
+ "model_type": "wav2vec2",
72
+ "no_mask_channel_overlap": false,
73
+ "no_mask_time_overlap": false,
74
+ "num_attention_heads": 12,
75
+ "num_codevector_groups": 2,
76
+ "num_codevectors_per_group": 320,
77
+ "num_conv_pos_embedding_groups": 16,
78
+ "num_conv_pos_embeddings": 128,
79
+ "num_feat_extract_layers": 7,
80
+ "num_hidden_layers": 12,
81
+ "num_negatives": 100,
82
+ "pad_token_id": 29,
83
+ "proj_codevector_dim": 256,
84
+ "torch_dtype": "float32",
85
+ "transformers_version": "4.11.3",
86
+ "use_weighted_layer_sum": false,
87
+ "vocab_size": 32
88
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4092458c130d33b60f2a9dda8e02e9153f9f4546d61e855641802b2ad8a15984
3
+ size 377670039
runs/Feb17_10-01-55_5fabccd8880d/1645092213.5924044/events.out.tfevents.1645092213.5fabccd8880d.84.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c69d3fc9d2071dda49dfdfa852764de3c964e397095458ab29405a9d9e2ea167
3
+ size 4519
runs/Feb17_10-01-55_5fabccd8880d/events.out.tfevents.1645092213.5fabccd8880d.84.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca5a0385d7c63bdfee10519a502607ae00a12d9d9b790a3a17fc20594bf8e244
3
+ size 8584
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1b07d1dd4cc6493def1bd7e8913100e68f098e1ae5794305fd1ea27af2e811b
3
+ size 2799