Hugging Face's logo

csikasote
/
wav2vec2-base-librispeech-model

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
libri10h
mms
Generated from Trainer
Model card Files Metrics Community
csikasote commited on
Commit
c7f2a84
·
verified ·
1 Parent(s): 396c8c8

Model save

Browse files
Files changed (2) hide show
  1. README.md +74 -48
  2. model.safetensors +1 -1
README.md CHANGED
@@ -3,9 +3,6 @@ library_name: transformers
3
  license: apache-2.0
4
  base_model: facebook/wav2vec2-base
5
  tags:
6
- - automatic-speech-recognition
7
- - libri10h
8
- - mms
9
  - generated_from_trainer
10
  metrics:
11
  - wer
@@ -19,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  # wav2vec2-base-librispeech-model
21
 
22
- This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the LIBRI10H - ENG dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.5788
25
- - Wer: 0.5170
26
 
27
  ## Model description
28
 
@@ -41,7 +38,7 @@ More information needed
41
  ### Training hyperparameters
42
 
43
  The following hyperparameters were used during training:
44
- - learning_rate: 5e-05
45
  - train_batch_size: 4
46
  - eval_batch_size: 4
47
  - seed: 42
@@ -50,53 +47,82 @@ The following hyperparameters were used during training:
50
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_steps: 500
53
- - training_steps: 20000
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer |
59
  |:-------------:|:-------:|:-----:|:---------------:|:------:|
60
- | 4.6193 | 1.4493 | 500 | 2.8892 | 1.0 |
61
- | 2.7664 | 2.8986 | 1000 | 2.3170 | 0.9991 |
62
- | 1.9103 | 4.3478 | 1500 | 1.4848 | 0.9327 |
63
- | 1.5308 | 5.7971 | 2000 | 1.2589 | 0.8841 |
64
- | 1.3511 | 7.2464 | 2500 | 1.1335 | 0.8555 |
65
- | 1.2321 | 8.6957 | 3000 | 1.0360 | 0.8070 |
66
- | 1.1402 | 10.1449 | 3500 | 0.9679 | 0.7786 |
67
- | 1.0625 | 11.5942 | 4000 | 0.9107 | 0.7560 |
68
- | 0.9972 | 13.0435 | 4500 | 0.8602 | 0.7238 |
69
- | 0.9427 | 14.4928 | 5000 | 0.8226 | 0.7057 |
70
- | 0.9031 | 15.9420 | 5500 | 0.7937 | 0.6869 |
71
- | 0.8577 | 17.3913 | 6000 | 0.7671 | 0.6696 |
72
- | 0.8252 | 18.8406 | 6500 | 0.7443 | 0.6570 |
73
- | 0.7945 | 20.2899 | 7000 | 0.7297 | 0.6482 |
74
- | 0.7668 | 21.7391 | 7500 | 0.7089 | 0.6295 |
75
- | 0.7387 | 23.1884 | 8000 | 0.7029 | 0.6244 |
76
- | 0.7196 | 24.6377 | 8500 | 0.6776 | 0.6096 |
77
- | 0.6943 | 26.0870 | 9000 | 0.6669 | 0.6032 |
78
- | 0.6764 | 27.5362 | 9500 | 0.6576 | 0.5930 |
79
- | 0.6577 | 28.9855 | 10000 | 0.6425 | 0.5797 |
80
- | 0.6436 | 30.4348 | 10500 | 0.6355 | 0.5767 |
81
- | 0.6264 | 31.8841 | 11000 | 0.6313 | 0.5709 |
82
- | 0.613 | 33.3333 | 11500 | 0.6237 | 0.5614 |
83
- | 0.604 | 34.7826 | 12000 | 0.6145 | 0.5595 |
84
- | 0.5863 | 36.2319 | 12500 | 0.6104 | 0.5537 |
85
- | 0.58 | 37.6812 | 13000 | 0.6109 | 0.5545 |
86
- | 0.5667 | 39.1304 | 13500 | 0.6045 | 0.5441 |
87
- | 0.5576 | 40.5797 | 14000 | 0.6008 | 0.5395 |
88
- | 0.5506 | 42.0290 | 14500 | 0.5925 | 0.5340 |
89
- | 0.5394 | 43.4783 | 15000 | 0.5933 | 0.5324 |
90
- | 0.5383 | 44.9275 | 15500 | 0.5908 | 0.5319 |
91
- | 0.5272 | 46.3768 | 16000 | 0.5875 | 0.5275 |
92
- | 0.5219 | 47.8261 | 16500 | 0.5849 | 0.5259 |
93
- | 0.5152 | 49.2754 | 17000 | 0.5878 | 0.5218 |
94
- | 0.5137 | 50.7246 | 17500 | 0.5834 | 0.5222 |
95
- | 0.5101 | 52.1739 | 18000 | 0.5830 | 0.5191 |
96
- | 0.5039 | 53.6232 | 18500 | 0.5816 | 0.5193 |
97
- | 0.5018 | 55.0725 | 19000 | 0.5798 | 0.5184 |
98
- | 0.4999 | 56.5217 | 19500 | 0.5806 | 0.5176 |
99
- | 0.4982 | 57.9710 | 20000 | 0.5789 | 0.5165 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
100
 
101
 
102
  ### Framework versions
 
3
  license: apache-2.0
4
  base_model: facebook/wav2vec2-base
5
  tags:
 
 
 
6
  - generated_from_trainer
7
  metrics:
8
  - wer
 
16
 
17
  # wav2vec2-base-librispeech-model
18
 
19
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.8155
22
+ - Wer: 0.3428
23
 
24
  ## Model description
25
 
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - learning_rate: 0.0003
42
  - train_batch_size: 4
43
  - eval_batch_size: 4
44
  - seed: 42
 
47
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
  - lr_scheduler_warmup_steps: 500
50
+ - num_epochs: 100.0
51
  - mixed_precision_training: Native AMP
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Wer |
56
  |:-------------:|:-------:|:-----:|:---------------:|:------:|
57
+ | 3.544 | 1.4493 | 500 | 1.7568 | 0.9820 |
58
+ | 1.42 | 2.8986 | 1000 | 1.0275 | 0.8168 |
59
+ | 1.0403 | 4.3478 | 1500 | 0.8305 | 0.7173 |
60
+ | 0.8574 | 5.7971 | 2000 | 0.7293 | 0.6649 |
61
+ | 0.7315 | 7.2464 | 2500 | 0.6632 | 0.6025 |
62
+ | 0.6389 | 8.6957 | 3000 | 0.6286 | 0.5695 |
63
+ | 0.5679 | 10.1449 | 3500 | 0.6102 | 0.5489 |
64
+ | 0.5085 | 11.5942 | 4000 | 0.5863 | 0.5215 |
65
+ | 0.4579 | 13.0435 | 4500 | 0.5661 | 0.4933 |
66
+ | 0.4097 | 14.4928 | 5000 | 0.5646 | 0.4823 |
67
+ | 0.382 | 15.9420 | 5500 | 0.5515 | 0.4644 |
68
+ | 0.3426 | 17.3913 | 6000 | 0.5585 | 0.4514 |
69
+ | 0.32 | 18.8406 | 6500 | 0.5598 | 0.4475 |
70
+ | 0.2926 | 20.2899 | 7000 | 0.6246 | 0.4587 |
71
+ | 0.2735 | 21.7391 | 7500 | 0.5887 | 0.4439 |
72
+ | 0.257 | 23.1884 | 8000 | 0.5978 | 0.4355 |
73
+ | 0.2409 | 24.6377 | 8500 | 0.5721 | 0.4215 |
74
+ | 0.2246 | 26.0870 | 9000 | 0.5979 | 0.4187 |
75
+ | 0.2139 | 27.5362 | 9500 | 0.6103 | 0.4145 |
76
+ | 0.2014 | 28.9855 | 10000 | 0.6436 | 0.4157 |
77
+ | 0.1917 | 30.4348 | 10500 | 0.6471 | 0.4188 |
78
+ | 0.184 | 31.8841 | 11000 | 0.6410 | 0.4068 |
79
+ | 0.1752 | 33.3333 | 11500 | 0.6426 | 0.4086 |
80
+ | 0.169 | 34.7826 | 12000 | 0.6633 | 0.4025 |
81
+ | 0.1612 | 36.2319 | 12500 | 0.6466 | 0.3968 |
82
+ | 0.1553 | 37.6812 | 13000 | 0.6573 | 0.3941 |
83
+ | 0.15 | 39.1304 | 13500 | 0.6989 | 0.3956 |
84
+ | 0.1442 | 40.5797 | 14000 | 0.7209 | 0.4062 |
85
+ | 0.1409 | 42.0290 | 14500 | 0.6950 | 0.3961 |
86
+ | 0.1356 | 43.4783 | 15000 | 0.6816 | 0.3863 |
87
+ | 0.134 | 44.9275 | 15500 | 0.6896 | 0.3867 |
88
+ | 0.1288 | 46.3768 | 16000 | 0.7073 | 0.3844 |
89
+ | 0.1263 | 47.8261 | 16500 | 0.7207 | 0.3836 |
90
+ | 0.1218 | 49.2754 | 17000 | 0.7430 | 0.3812 |
91
+ | 0.1217 | 50.7246 | 17500 | 0.7588 | 0.3831 |
92
+ | 0.1183 | 52.1739 | 18000 | 0.7478 | 0.3813 |
93
+ | 0.113 | 53.6232 | 18500 | 0.7269 | 0.3779 |
94
+ | 0.1109 | 55.0725 | 19000 | 0.7117 | 0.3735 |
95
+ | 0.1102 | 56.5217 | 19500 | 0.7532 | 0.3689 |
96
+ | 0.1084 | 57.9710 | 20000 | 0.7608 | 0.3704 |
97
+ | 0.1042 | 59.4203 | 20500 | 0.7571 | 0.3677 |
98
+ | 0.1048 | 60.8696 | 21000 | 0.7745 | 0.3683 |
99
+ | 0.1005 | 62.3188 | 21500 | 0.7845 | 0.3712 |
100
+ | 0.1006 | 63.7681 | 22000 | 0.7633 | 0.3664 |
101
+ | 0.0976 | 65.2174 | 22500 | 0.7721 | 0.3639 |
102
+ | 0.096 | 66.6667 | 23000 | 0.7659 | 0.3643 |
103
+ | 0.0938 | 68.1159 | 23500 | 0.7658 | 0.3620 |
104
+ | 0.0933 | 69.5652 | 24000 | 0.7692 | 0.3579 |
105
+ | 0.092 | 71.0145 | 24500 | 0.7785 | 0.3625 |
106
+ | 0.089 | 72.4638 | 25000 | 0.7845 | 0.3615 |
107
+ | 0.088 | 73.9130 | 25500 | 0.7973 | 0.3586 |
108
+ | 0.0862 | 75.3623 | 26000 | 0.7806 | 0.3576 |
109
+ | 0.0851 | 76.8116 | 26500 | 0.7947 | 0.3583 |
110
+ | 0.0846 | 78.2609 | 27000 | 0.7802 | 0.3526 |
111
+ | 0.0809 | 79.7101 | 27500 | 0.8093 | 0.3532 |
112
+ | 0.0813 | 81.1594 | 28000 | 0.8237 | 0.3572 |
113
+ | 0.0785 | 82.6087 | 28500 | 0.8130 | 0.3533 |
114
+ | 0.0799 | 84.0580 | 29000 | 0.7958 | 0.3511 |
115
+ | 0.0784 | 85.5072 | 29500 | 0.8108 | 0.3507 |
116
+ | 0.0767 | 86.9565 | 30000 | 0.8208 | 0.3511 |
117
+ | 0.0742 | 88.4058 | 30500 | 0.8270 | 0.3501 |
118
+ | 0.0746 | 89.8551 | 31000 | 0.8121 | 0.3459 |
119
+ | 0.073 | 91.3043 | 31500 | 0.8151 | 0.3485 |
120
+ | 0.0725 | 92.7536 | 32000 | 0.8265 | 0.3477 |
121
+ | 0.0717 | 94.2029 | 32500 | 0.8173 | 0.3446 |
122
+ | 0.0709 | 95.6522 | 33000 | 0.8135 | 0.3434 |
123
+ | 0.0704 | 97.1014 | 33500 | 0.8179 | 0.3431 |
124
+ | 0.0699 | 98.5507 | 34000 | 0.8134 | 0.3427 |
125
+ | 0.0691 | 100.0 | 34500 | 0.8155 | 0.3428 |
126
 
127
 
128
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6fcc9ac1bbc1fd9ca0e48dec42b75713d8c36b779c86b4fc2de03ae0e0fa6892
3
  size 415620972
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d21d6ea41f74cee39cc0861b3b545646cf734879ff7ee872ae09b8858c0aefd5
3
  size 415620972