Hugging Face's logo

addy88
/
wav2vec2-base-timit-english

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Inference Endpoints
Model card Files Metrics Community
addy88 commited on
Commit
27caf1a
•
1 Parent(s): 47912dc

add tokenizer

Browse files
Files changed (25) 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/scaler.pt +3 -0
  7. checkpoint-3500/scheduler.pt +3 -0
  8. checkpoint-3500/trainer_state.json +121 -0
  9. checkpoint-3500/training_args.bin +3 -0
  10. checkpoint-4000/config.json +88 -0
  11. checkpoint-4000/optimizer.pt +3 -0
  12. checkpoint-4000/preprocessor_config.json +9 -0
  13. checkpoint-4000/pytorch_model.bin +3 -0
  14. checkpoint-4000/rng_state.pth +3 -0
  15. checkpoint-4000/scaler.pt +3 -0
  16. checkpoint-4000/scheduler.pt +3 -0
  17. checkpoint-4000/trainer_state.json +136 -0
  18. checkpoint-4000/training_args.bin +3 -0
  19. config.json +88 -0
  20. preprocessor_config.json +9 -0
  21. pytorch_model.bin +3 -0
  22. runs/Dec09_05-45-29_0e7932201914/1639029117.8821757/events.out.tfevents.1639029117.0e7932201914.179.1 +3 -0
  23. runs/Dec09_05-45-29_0e7932201914/events.out.tfevents.1639029117.0e7932201914.179.0 +3 -0
  24. training_args.bin +3 -0
  25. vocab.json +1 -1
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": 0.05,
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:a2a2dbecb1321edf0372aa63830fa1488fb60c28049aee8a4a7dfe63ece94b00
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:c10e2f151b12f4b340e1cb5d64fc744477af997c8e450ed7efd2ba9d1002fba4
3
+ size 377670039
checkpoint-3500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40ae2265452089af3437c7592d2b438a5743c11a3132585f071fc4e9036a0cbb
3
+ size 14567
checkpoint-3500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c1c179922cf50137f4f01691951993a4d476691e00d342a46b702a263b34f9f
3
+ size 559
checkpoint-3500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31863c9168f8bac807fcd1f7aa5fdea90fd5bf14712c0abaa0dec9e207b249e2
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": 4.97e-05,
13
+ "loss": 3.8466,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.45,
18
+ "eval_loss": 2.4666473865509033,
19
+ "eval_runtime": 66.0325,
20
+ "eval_samples_per_second": 25.442,
21
+ "eval_steps_per_second": 3.18,
22
+ "eval_wer": 1.0,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 6.9,
27
+ "learning_rate": 9.970000000000001e-05,
28
+ "loss": 0.9403,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 6.9,
33
+ "eval_loss": 0.4411664605140686,
34
+ "eval_runtime": 66.1692,
35
+ "eval_samples_per_second": 25.389,
36
+ "eval_steps_per_second": 3.174,
37
+ "eval_wer": 0.39928330232237613,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 10.34,
42
+ "learning_rate": 8.516417910447761e-05,
43
+ "loss": 0.2824,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 10.34,
48
+ "eval_loss": 0.3629843592643738,
49
+ "eval_runtime": 65.8604,
50
+ "eval_samples_per_second": 25.508,
51
+ "eval_steps_per_second": 3.189,
52
+ "eval_wer": 0.3318172420922059,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 13.79,
57
+ "learning_rate": 7.023880597014926e-05,
58
+ "loss": 0.1639,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 13.79,
63
+ "eval_loss": 0.3536781072616577,
64
+ "eval_runtime": 65.6765,
65
+ "eval_samples_per_second": 25.58,
66
+ "eval_steps_per_second": 3.197,
67
+ "eval_wer": 0.3030115085107849,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 17.24,
72
+ "learning_rate": 5.53134328358209e-05,
73
+ "loss": 0.1251,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 17.24,
78
+ "eval_loss": 0.39447325468063354,
79
+ "eval_runtime": 66.1834,
80
+ "eval_samples_per_second": 25.384,
81
+ "eval_steps_per_second": 3.173,
82
+ "eval_wer": 0.2973606229756736,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 20.69,
87
+ "learning_rate": 4.0388059701492536e-05,
88
+ "loss": 0.1075,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 20.69,
93
+ "eval_loss": 0.3765473961830139,
94
+ "eval_runtime": 66.4767,
95
+ "eval_samples_per_second": 25.272,
96
+ "eval_steps_per_second": 3.159,
97
+ "eval_wer": 0.29370822134932123,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 24.14,
102
+ "learning_rate": 2.546268656716418e-05,
103
+ "loss": 0.0831,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 24.14,
108
+ "eval_loss": 0.3624289631843567,
109
+ "eval_runtime": 66.6948,
110
+ "eval_samples_per_second": 25.189,
111
+ "eval_steps_per_second": 3.149,
112
+ "eval_wer": 0.2816484046585349,
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 sha256:6643b08a25ae22ed55383c349589e3a74c18fbd783c63dca0ef6a60a259de663
3
+ size 2863
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": 0.05,
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 sha256:57b6dd8ea40f41f9af4f12244a2d28ac16f544c46a19120dbea4304f27122c02
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:fdd50aa06b7b1fbd62a8c0d8883d2df4aec963849ae924ef27671d99154ca320
3
+ size 377670039
checkpoint-4000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:659592d6ec69501356c0ac1fb3b3b8b7e01ffa731fe678fb2077a3db80e62a4d
3
+ size 14567
checkpoint-4000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fe4a4d1db499cb8dcc9f4dffff5f8a478c41f87039186872835823d1631b3d5
3
+ size 559
checkpoint-4000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:972beee84ef9a2830f2e2565c7169d81f2fec6e815d9f1edca59b246e4fcbce3
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": 4.97e-05,
13
+ "loss": 3.8466,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.45,
18
+ "eval_loss": 2.4666473865509033,
19
+ "eval_runtime": 66.0325,
20
+ "eval_samples_per_second": 25.442,
21
+ "eval_steps_per_second": 3.18,
22
+ "eval_wer": 1.0,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 6.9,
27
+ "learning_rate": 9.970000000000001e-05,
28
+ "loss": 0.9403,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 6.9,
33
+ "eval_loss": 0.4411664605140686,
34
+ "eval_runtime": 66.1692,
35
+ "eval_samples_per_second": 25.389,
36
+ "eval_steps_per_second": 3.174,
37
+ "eval_wer": 0.39928330232237613,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 10.34,
42
+ "learning_rate": 8.516417910447761e-05,
43
+ "loss": 0.2824,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 10.34,
48
+ "eval_loss": 0.3629843592643738,
49
+ "eval_runtime": 65.8604,
50
+ "eval_samples_per_second": 25.508,
51
+ "eval_steps_per_second": 3.189,
52
+ "eval_wer": 0.3318172420922059,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 13.79,
57
+ "learning_rate": 7.023880597014926e-05,
58
+ "loss": 0.1639,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 13.79,
63
+ "eval_loss": 0.3536781072616577,
64
+ "eval_runtime": 65.6765,
65
+ "eval_samples_per_second": 25.58,
66
+ "eval_steps_per_second": 3.197,
67
+ "eval_wer": 0.3030115085107849,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 17.24,
72
+ "learning_rate": 5.53134328358209e-05,
73
+ "loss": 0.1251,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 17.24,
78
+ "eval_loss": 0.39447325468063354,
79
+ "eval_runtime": 66.1834,
80
+ "eval_samples_per_second": 25.384,
81
+ "eval_steps_per_second": 3.173,
82
+ "eval_wer": 0.2973606229756736,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 20.69,
87
+ "learning_rate": 4.0388059701492536e-05,
88
+ "loss": 0.1075,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 20.69,
93
+ "eval_loss": 0.3765473961830139,
94
+ "eval_runtime": 66.4767,
95
+ "eval_samples_per_second": 25.272,
96
+ "eval_steps_per_second": 3.159,
97
+ "eval_wer": 0.29370822134932123,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 24.14,
102
+ "learning_rate": 2.546268656716418e-05,
103
+ "loss": 0.0831,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 24.14,
108
+ "eval_loss": 0.3624289631843567,
109
+ "eval_runtime": 66.6948,
110
+ "eval_samples_per_second": 25.189,
111
+ "eval_steps_per_second": 3.149,
112
+ "eval_wer": 0.2816484046585349,
113
+ "step": 3500
114
+ },
115
+ {
116
+ "epoch": 27.59,
117
+ "learning_rate": 1.0537313432835822e-05,
118
+ "loss": 0.0639,
119
+ "step": 4000
120
+ },
121
+ {
122
+ "epoch": 27.59,
123
+ "eval_loss": 0.39012792706489563,
124
+ "eval_runtime": 66.5439,
125
+ "eval_samples_per_second": 25.247,
126
+ "eval_steps_per_second": 3.156,
127
+ "eval_wer": 0.28157949142030186,
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 sha256:6643b08a25ae22ed55383c349589e3a74c18fbd783c63dca0ef6a60a259de663
3
+ size 2863
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": 0.05,
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:e2d54f606c17817efcf788545b7d46d5c61115ac4df7e906208e673840ea579e
3
+ size 377670039
runs/Dec09_05-45-29_0e7932201914/1639029117.8821757/events.out.tfevents.1639029117.0e7932201914.179.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e498d96a3b3881350a1f83bb7542fec5c6fc2f5d8adb6ae1123d953befcba4c7
3
+ size 4546
runs/Dec09_05-45-29_0e7932201914/events.out.tfevents.1639029117.0e7932201914.179.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b713c0c969477bbe82116ec781dad9683c0d45ec395f0963cf206f1d93cc5213
3
+ size 8611
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6643b08a25ae22ed55383c349589e3a74c18fbd783c63dca0ef6a60a259de663
3
+ size 2863
vocab.json CHANGED
@@ -1 +1 @@
1
- {"z": 0, "f": 1, "q": 2, "x": 3, "n": 4, "d": 5, "r": 6, "a": 7, "m": 9, "t": 10, "h": 11, "l": 12, "g": 13, "k": 14, "p": 15, "e": 16, "i": 17, "j": 18, "v": 19, "u": 20, "c": 21, "w": 22, "o": 23, "y": 24, "'": 25, "s": 26, "b": 27, "|": 8, "[UNK]": 28, "[PAD]": 29}
1
+ {"m": 0, "v": 1, "t": 2, "r": 3, "n": 5, "d": 6, "k": 7, "b": 8, "y": 9, "i": 10, "g": 11, "j": 12, "f": 13, "x": 14, "a": 15, "p": 16, "u": 17, "'": 18, "c": 19, "o": 20, "l": 21, "e": 22, "z": 23, "q": 24, "w": 25, "h": 26, "s": 27, "|": 4, "[UNK]": 28, "[PAD]": 29}