Hugging Face's logo

patrickvonplaten
/
wavlm-libri-clean-100h-large

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wavlm
librispeech_asr
Generated from Trainer
wavlm_libri_finetune
Inference Endpoints
Model card Files Metrics Community
3
patrickvonplaten commited on
Commit
6ca1c28
•
1 Parent(s): a1b7ace

up

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +72 -0
  2. added_tokens.json +1 -0
  3. all_results.json +14 -0
  4. checkpoint-1500/config.json +98 -0
  5. checkpoint-1500/optimizer.pt +3 -0
  6. checkpoint-1500/preprocessor_config.json +9 -0
  7. checkpoint-1500/pytorch_model.bin +3 -0
  8. checkpoint-1500/rng_state_0.pth +3 -0
  9. checkpoint-1500/rng_state_1.pth +3 -0
  10. checkpoint-1500/rng_state_2.pth +3 -0
  11. checkpoint-1500/rng_state_3.pth +3 -0
  12. checkpoint-1500/rng_state_4.pth +3 -0
  13. checkpoint-1500/rng_state_5.pth +3 -0
  14. checkpoint-1500/rng_state_6.pth +3 -0
  15. checkpoint-1500/rng_state_7.pth +3 -0
  16. checkpoint-1500/scaler.pt +3 -0
  17. checkpoint-1500/scheduler.pt +3 -0
  18. checkpoint-1500/trainer_state.json +0 -0
  19. checkpoint-1500/training_args.bin +3 -0
  20. checkpoint-2000/config.json +98 -0
  21. checkpoint-2000/optimizer.pt +3 -0
  22. checkpoint-2000/preprocessor_config.json +9 -0
  23. checkpoint-2000/pytorch_model.bin +3 -0
  24. checkpoint-2000/rng_state_0.pth +3 -0
  25. checkpoint-2000/rng_state_1.pth +3 -0
  26. checkpoint-2000/rng_state_2.pth +3 -0
  27. checkpoint-2000/rng_state_3.pth +3 -0
  28. checkpoint-2000/rng_state_4.pth +3 -0
  29. checkpoint-2000/rng_state_5.pth +3 -0
  30. checkpoint-2000/rng_state_6.pth +3 -0
  31. checkpoint-2000/rng_state_7.pth +3 -0
  32. checkpoint-2000/scaler.pt +3 -0
  33. checkpoint-2000/scheduler.pt +3 -0
  34. checkpoint-2000/trainer_state.json +0 -0
  35. checkpoint-2000/training_args.bin +3 -0
  36. checkpoint-2500/config.json +98 -0
  37. checkpoint-2500/optimizer.pt +3 -0
  38. checkpoint-2500/preprocessor_config.json +9 -0
  39. checkpoint-2500/pytorch_model.bin +3 -0
  40. checkpoint-2500/rng_state_0.pth +3 -0
  41. checkpoint-2500/rng_state_1.pth +3 -0
  42. checkpoint-2500/rng_state_2.pth +3 -0
  43. checkpoint-2500/rng_state_3.pth +3 -0
  44. checkpoint-2500/rng_state_4.pth +3 -0
  45. checkpoint-2500/rng_state_5.pth +3 -0
  46. checkpoint-2500/rng_state_6.pth +3 -0
  47. checkpoint-2500/rng_state_7.pth +3 -0
  48. checkpoint-2500/scaler.pt +3 -0
  49. checkpoint-2500/scheduler.pt +3 -0
  50. checkpoint-2500/trainer_state.json +0 -0
README.md ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - automatic-speech-recognition
4
+ - librispeech_asr
5
+ - generated_from_trainer
6
+ model-index:
7
+ - name: wavlm-librispeech-clean-100h-dist
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ # wavlm-librispeech-clean-100h-dist
15
+
16
+ This model is a fine-tuned version of [microsoft/wavlm-large](https://huggingface.co/microsoft/wavlm-large) on the LIBRISPEECH_ASR - CLEAN dataset.
17
+ It achieves the following results on the evaluation set:
18
+ - Loss: 0.0601
19
+ - Wer: 0.0491
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 0.0003
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 2
41
+ - seed: 42
42
+ - distributed_type: multi-GPU
43
+ - num_devices: 8
44
+ - gradient_accumulation_steps: 2
45
+ - total_train_batch_size: 32
46
+ - total_eval_batch_size: 16
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_steps: 500
50
+ - num_epochs: 3.0
51
+ - mixed_precision_training: Native AMP
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
56
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
57
+ | 0.8069 | 0.34 | 300 | 0.7510 | 0.5809 |
58
+ | 0.2483 | 0.67 | 600 | 0.2023 | 0.1929 |
59
+ | 0.1033 | 1.01 | 900 | 0.1123 | 0.1028 |
60
+ | 0.0742 | 1.35 | 1200 | 0.0858 | 0.0771 |
61
+ | 0.057 | 1.68 | 1500 | 0.0722 | 0.0663 |
62
+ | 0.0421 | 2.02 | 1800 | 0.0682 | 0.0582 |
63
+ | 0.0839 | 2.35 | 2100 | 0.0630 | 0.0534 |
64
+ | 0.0307 | 2.69 | 2400 | 0.0603 | 0.0508 |
65
+
66
+
67
+ ### Framework versions
68
+
69
+ - Transformers 4.15.0.dev0
70
+ - Pytorch 1.9.0+cu111
71
+ - Datasets 1.16.2.dev0
72
+ - Tokenizers 0.10.3
added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"<s>": 29, "</s>": 30}
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.06008534133434296,
4
+ "eval_runtime": 60.484,
5
+ "eval_samples": 2703,
6
+ "eval_samples_per_second": 44.69,
7
+ "eval_steps_per_second": 2.794,
8
+ "eval_wer": 0.04911584132936289,
9
+ "train_loss": 0.48771899742295655,
10
+ "train_runtime": 3791.9372,
11
+ "train_samples": 28539,
12
+ "train_samples_per_second": 22.579,
13
+ "train_steps_per_second": 0.706
14
+ }
checkpoint-1500/config.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/wavlm-large",
3
+ "activation_dropout": 0.0,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "WavLMForCTC"
10
+ ],
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": false,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "mean",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": true,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_dropout": 0.0,
51
+ "feat_extract_norm": "layer",
52
+ "feat_proj_dropout": 0.0,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.0,
57
+ "hidden_size": 1024,
58
+ "initializer_range": 0.02,
59
+ "intermediate_size": 4096,
60
+ "layer_norm_eps": 1e-05,
61
+ "layerdrop": 0.0,
62
+ "mask_channel_length": 10,
63
+ "mask_channel_min_space": 1,
64
+ "mask_channel_other": 0.0,
65
+ "mask_channel_prob": 0.0,
66
+ "mask_channel_selection": "static",
67
+ "mask_feature_length": 10,
68
+ "mask_feature_min_masks": 0,
69
+ "mask_feature_prob": 0.0,
70
+ "mask_time_length": 10,
71
+ "mask_time_min_masks": 2,
72
+ "mask_time_min_space": 1,
73
+ "mask_time_other": 0.0,
74
+ "mask_time_prob": 0.05,
75
+ "mask_time_selection": "static",
76
+ "max_bucket_distance": 800,
77
+ "model_type": "wavlm",
78
+ "num_adapter_layers": 3,
79
+ "num_attention_heads": 16,
80
+ "num_buckets": 320,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_ctc_classes": 80,
86
+ "num_feat_extract_layers": 7,
87
+ "num_hidden_layers": 24,
88
+ "num_negatives": 100,
89
+ "output_hidden_size": 1024,
90
+ "pad_token_id": 28,
91
+ "proj_codevector_dim": 768,
92
+ "replace_prob": 0.5,
93
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
94
+ "torch_dtype": "float32",
95
+ "transformers_version": "4.15.0.dev0",
96
+ "use_weighted_layer_sum": false,
97
+ "vocab_size": 31
98
+ }
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb193a996ede6126a7d71fcbddcad44727a203e49e7b6138bade6786b5620430
3
+ size 2490498221
checkpoint-1500/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": true,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-1500/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:265036c53afbf09c9b739a005b83015de77798a518fe2b22007f99029f4f6a25
3
+ size 1262130157
checkpoint-1500/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bda153883946f3813bbd5c83628fdd0ffa30c6dd5743299c5708c832d555a8e6
3
+ size 14567
checkpoint-1500/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d40cc87b842b7071d8ed6058efa5f5ea399117b7ddd8b86c10f4a739725a108
3
+ size 14567
checkpoint-1500/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fabcce483b93608a7733dc60fd5997ba6d4460cc10c4669b05e30b00968923d
3
+ size 14503
checkpoint-1500/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1aeb77b213c0853d1b31a9ab26d743850336c2581c4a71bd2386c66fe21f8e4c
3
+ size 14503
checkpoint-1500/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:643588e6dafa6a4999c3062be19154e6952f589053d9e7da8683d0592790b4fd
3
+ size 14503
checkpoint-1500/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:400845dc8dd5421a893730d9ba026799a1c8819f400d72848117fe1331b4449a
3
+ size 14567
checkpoint-1500/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33ec89e63c36a2d101d43741b22ac4c77ad2c5f4a40c72f66b61c95c025358a6
3
+ size 14567
checkpoint-1500/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43762edb1856f1b1d8b7cea662e408a1fd2877a5878c5b16bb509529f901645b
3
+ size 14503
checkpoint-1500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82596ae7715fd1bbe7a2aeefffd658174bb33ef2f604c549bc24c13118a2b369
3
+ size 559
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbb5f85c51b9aac6ec915dc6207c7d1c2857ac9dfd7bf94c9c57c093dddf094f
3
+ size 623
checkpoint-1500/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16a69eb1dc7d0fd6fbada6d6bbfd9e4b781eb0c64524b8c51d51d08eeedb0251
3
+ size 2991
checkpoint-2000/config.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/wavlm-large",
3
+ "activation_dropout": 0.0,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "WavLMForCTC"
10
+ ],
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": false,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "mean",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": true,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_dropout": 0.0,
51
+ "feat_extract_norm": "layer",
52
+ "feat_proj_dropout": 0.0,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.0,
57
+ "hidden_size": 1024,
58
+ "initializer_range": 0.02,
59
+ "intermediate_size": 4096,
60
+ "layer_norm_eps": 1e-05,
61
+ "layerdrop": 0.0,
62
+ "mask_channel_length": 10,
63
+ "mask_channel_min_space": 1,
64
+ "mask_channel_other": 0.0,
65
+ "mask_channel_prob": 0.0,
66
+ "mask_channel_selection": "static",
67
+ "mask_feature_length": 10,
68
+ "mask_feature_min_masks": 0,
69
+ "mask_feature_prob": 0.0,
70
+ "mask_time_length": 10,
71
+ "mask_time_min_masks": 2,
72
+ "mask_time_min_space": 1,
73
+ "mask_time_other": 0.0,
74
+ "mask_time_prob": 0.05,
75
+ "mask_time_selection": "static",
76
+ "max_bucket_distance": 800,
77
+ "model_type": "wavlm",
78
+ "num_adapter_layers": 3,
79
+ "num_attention_heads": 16,
80
+ "num_buckets": 320,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_ctc_classes": 80,
86
+ "num_feat_extract_layers": 7,
87
+ "num_hidden_layers": 24,
88
+ "num_negatives": 100,
89
+ "output_hidden_size": 1024,
90
+ "pad_token_id": 28,
91
+ "proj_codevector_dim": 768,
92
+ "replace_prob": 0.5,
93
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
94
+ "torch_dtype": "float32",
95
+ "transformers_version": "4.15.0.dev0",
96
+ "use_weighted_layer_sum": false,
97
+ "vocab_size": 31
98
+ }
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:792d60ae639124d8d3d554f24f10932e3d5703720b7bd5bf81e12d3611b7916f
3
+ size 2490498221
checkpoint-2000/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": true,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-2000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18644d86f64f66264f66f70bf1dcd150ca183565a4a71cec05d35aaba1a77ccf
3
+ size 1262130157
checkpoint-2000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e5cac74d16e2c1e8ae42ad04766e7137b4044fbf52cf500340ce4b7220107c4
3
+ size 14567
checkpoint-2000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42aa9cb5054ebc48a7dae97cc7d9fe0c656e67c33e00b6c45ebf7e6449f0f31c
3
+ size 14567
checkpoint-2000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6252653fc118cc510c4ce0d356e008cb7c726a3c08e37b612e59afa0aaf51522
3
+ size 14503
checkpoint-2000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3767dd70edaf5e4b30016c847a8b2335b20bb0a2b4bfed7a2d3b1666fec128
3
+ size 14567
checkpoint-2000/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd81bcb00bac8a8a80af75b0484f3a7b891eb07ee09bfb13bb6a10fa476b4dca
3
+ size 14503
checkpoint-2000/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b921baebec423943fff3002bb46254ae002e83077efcb41722de905ddb8e8ed
3
+ size 14567
checkpoint-2000/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5af27a84ac1ad4b303d98e92e10cb9f523e86f4958f93de57c3640741b744da1
3
+ size 14567
checkpoint-2000/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34a488cd1dcca9e817909776619c372b3a7dcbf030bb8ef8b2ef24755333b508
3
+ size 14567
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:563f36c9bcf21bba6e269672f67bbe08e365f3b8a7797320cf86737ee95c1eae
3
+ size 559
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32bd226c77cc3f4d4280ddda0722570db3238ef7d04d7a310c4f52523d39b2fc
3
+ size 623
checkpoint-2000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16a69eb1dc7d0fd6fbada6d6bbfd9e4b781eb0c64524b8c51d51d08eeedb0251
3
+ size 2991
checkpoint-2500/config.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/wavlm-large",
3
+ "activation_dropout": 0.0,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "WavLMForCTC"
10
+ ],
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": false,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "mean",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": true,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_dropout": 0.0,
51
+ "feat_extract_norm": "layer",
52
+ "feat_proj_dropout": 0.0,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.0,
57
+ "hidden_size": 1024,
58
+ "initializer_range": 0.02,
59
+ "intermediate_size": 4096,
60
+ "layer_norm_eps": 1e-05,
61
+ "layerdrop": 0.0,
62
+ "mask_channel_length": 10,
63
+ "mask_channel_min_space": 1,
64
+ "mask_channel_other": 0.0,
65
+ "mask_channel_prob": 0.0,
66
+ "mask_channel_selection": "static",
67
+ "mask_feature_length": 10,
68
+ "mask_feature_min_masks": 0,
69
+ "mask_feature_prob": 0.0,
70
+ "mask_time_length": 10,
71
+ "mask_time_min_masks": 2,
72
+ "mask_time_min_space": 1,
73
+ "mask_time_other": 0.0,
74
+ "mask_time_prob": 0.05,
75
+ "mask_time_selection": "static",
76
+ "max_bucket_distance": 800,
77
+ "model_type": "wavlm",
78
+ "num_adapter_layers": 3,
79
+ "num_attention_heads": 16,
80
+ "num_buckets": 320,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_ctc_classes": 80,
86
+ "num_feat_extract_layers": 7,
87
+ "num_hidden_layers": 24,
88
+ "num_negatives": 100,
89
+ "output_hidden_size": 1024,
90
+ "pad_token_id": 28,
91
+ "proj_codevector_dim": 768,
92
+ "replace_prob": 0.5,
93
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
94
+ "torch_dtype": "float32",
95
+ "transformers_version": "4.15.0.dev0",
96
+ "use_weighted_layer_sum": false,
97
+ "vocab_size": 31
98
+ }
checkpoint-2500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a073dceb448668286238262eb5b2734e2434fccfaea38ac6e6ebb9cdff1599a
3
+ size 2490498221
checkpoint-2500/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": true,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-2500/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c778cd69a78d1fff0973971b4db5c3fcc7f08a8ab440b69e82212df90bfcbd3
3
+ size 1262130157
checkpoint-2500/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b2f67342d205d1b1bd606ac7357996f81944f198db55622487f27aa249bc5db
3
+ size 14567
checkpoint-2500/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b79e368239f883cc5b7958ecf0e851ac82f288e39815b19dd44b765b03ca185
3
+ size 14567
checkpoint-2500/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb6a9eb12efebcbe405aec5c8f22280cb58ad221c2eefd1d141f082224b6d55
3
+ size 14567
checkpoint-2500/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e6992d94ab3372b8263e8f724bd2e72e00e437b7a56e8dd63bb22bfc78a8787
3
+ size 14567
checkpoint-2500/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d74c20eca09f431b90ca9ca38759fb03f6ba4d4798b4161848b23cb0b68fd0fa
3
+ size 14567
checkpoint-2500/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ae0771316fc20fe519b53a63405b9043d36931088bf9954d96540a87fcd7e8b
3
+ size 14567
checkpoint-2500/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfb3c971ed248d8105e74fece3ad8442c75e206b15a4e8a92405889ef7e50e5c
3
+ size 14567
checkpoint-2500/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1286e0cba26484f3cd66e7ddbf24ce218fed723a2038a1c8aa7c6ce18426695f
3
+ size 14503
checkpoint-2500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8684a2ff80379dbf940dcb0062b2a4c72808ed76eca555dd9783d94320742aba
3
+ size 559
checkpoint-2500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f408d4fac841a378404b3ea6ea1a42643615129f25cd440d86b0577f06da6ce1
3
+ size 623
checkpoint-2500/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff