Hugging Face's logo

ctaguchi
/
ssc-aln-model

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
 Community
ctaguchi commited on
Commit
a842f9c
·
verified ·
1 Parent(s): 79b25fc

Model save

Browse files
Files changed (2) hide show
  1. README.md +55 -57
  2. model.safetensors +1 -1
README.md CHANGED
@@ -1,7 +1,5 @@
1
  ---
2
  library_name: transformers
3
- license: apache-2.0
4
- base_model: facebook/wav2vec2-xls-r-300m
5
  tags:
6
  - generated_from_trainer
7
  metrics:
@@ -16,11 +14,11 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # ssc-aln-model
18
 
19
- This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 2.2203
22
- - Cer: 0.7556
23
- - Wer: 0.9997
24
 
25
  ## Model description
26
 
@@ -40,69 +38,69 @@ More information needed
40
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 0.0003
43
- - train_batch_size: 8
44
  - eval_batch_size: 8
45
  - seed: 42
46
  - gradient_accumulation_steps: 2
47
- - total_train_batch_size: 16
48
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
  - lr_scheduler_type: linear
50
  - lr_scheduler_warmup_steps: 100
51
- - num_epochs: 10
52
  - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
57
  |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
58
- | 6.7941 | 0.2053 | 100 | 3.2691 | 0.9971 | 1.0 |
59
- | 3.4406 | 0.4107 | 200 | 3.1704 | 0.9972 | 1.0 |
60
- | 3.5811 | 0.6160 | 300 | 3.1971 | 0.9972 | 1.0 |
61
- | 3.4913 | 0.8214 | 400 | 3.1145 | 0.9972 | 1.0 |
62
- | 3.4276 | 1.0267 | 500 | 3.1228 | 0.9972 | 1.0 |
63
- | 3.4187 | 1.2320 | 600 | 3.1975 | 0.9972 | 1.0 |
64
- | 3.4298 | 1.4374 | 700 | 3.0807 | 0.9972 | 1.0 |
65
- | 3.4501 | 1.6427 | 800 | 3.0874 | 0.9972 | 1.0 |
66
- | 3.3777 | 1.8480 | 900 | 3.1311 | 0.9972 | 1.0 |
67
- | 3.3771 | 2.0534 | 1000 | 3.0617 | 0.9972 | 1.0 |
68
- | 3.3915 | 2.2587 | 1100 | 3.0929 | 0.9972 | 1.0 |
69
- | 3.4002 | 2.4641 | 1200 | 3.1512 | 0.9972 | 1.0 |
70
- | 3.3626 | 2.6694 | 1300 | 3.0500 | 0.9972 | 1.0 |
71
- | 3.3615 | 2.8747 | 1400 | 3.0784 | 0.9972 | 1.0 |
72
- | 3.3876 | 3.0801 | 1500 | 3.0095 | 0.9972 | 1.0 |
73
- | 3.3524 | 3.2854 | 1600 | 3.0289 | 0.9972 | 1.0 |
74
- | 3.3199 | 3.4908 | 1700 | 3.0130 | 0.9972 | 1.0 |
75
- | 3.3461 | 3.6961 | 1800 | 3.0453 | 0.9972 | 1.0 |
76
- | 3.3412 | 3.9014 | 1900 | 3.0517 | 0.9972 | 1.0 |
77
- | 3.3504 | 4.1068 | 2000 | 3.1052 | 0.9972 | 1.0 |
78
- | 3.2621 | 4.3121 | 2100 | 3.1311 | 0.9971 | 1.0 |
79
- | 3.3126 | 4.5175 | 2200 | 3.0152 | 0.9971 | 1.0 |
80
- | 3.2792 | 4.7228 | 2300 | 3.0300 | 0.9972 | 1.0 |
81
- | 3.1871 | 4.9281 | 2400 | 3.1724 | 0.9972 | 1.0 |
82
- | 3.193 | 5.1335 | 2500 | 3.0378 | 0.9972 | 1.0 |
83
- | 3.2125 | 5.3388 | 2600 | 2.9703 | 0.9971 | 1.0 |
84
- | 3.151 | 5.5441 | 2700 | 2.9920 | 0.9972 | 1.0 |
85
- | 3.0805 | 5.7495 | 2800 | 2.9865 | 0.9972 | 1.0 |
86
- | 3.0694 | 5.9548 | 2900 | 2.8673 | 0.9972 | 1.0 |
87
- | 3.0316 | 6.1602 | 3000 | 2.9323 | 0.9725 | 0.9925 |
88
- | 2.9225 | 6.3655 | 3100 | 2.7120 | 0.9692 | 0.9987 |
89
- | 2.9326 | 6.5708 | 3200 | 2.6417 | 0.9886 | 0.9997 |
90
- | 2.9268 | 6.7762 | 3300 | 2.6350 | 0.9948 | 1.0 |
91
- | 2.8654 | 6.9815 | 3400 | 2.5573 | 0.9885 | 0.9989 |
92
- | 2.795 | 7.1869 | 3500 | 2.5040 | 0.9134 | 0.9975 |
93
- | 2.7497 | 7.3922 | 3600 | 2.4582 | 0.9732 | 1.0 |
94
- | 2.7869 | 7.5975 | 3700 | 2.3806 | 0.8967 | 1.0 |
95
- | 2.7088 | 7.8029 | 3800 | 2.4244 | 0.8773 | 0.9999 |
96
- | 2.7059 | 8.0082 | 3900 | 2.3560 | 0.8910 | 0.9999 |
97
- | 2.635 | 8.2136 | 4000 | 2.3311 | 0.8075 | 0.9994 |
98
- | 2.6094 | 8.4189 | 4100 | 2.3087 | 0.8328 | 0.9999 |
99
- | 2.6053 | 8.6242 | 4200 | 2.2820 | 0.8365 | 0.9997 |
100
- | 2.6151 | 8.8296 | 4300 | 2.2771 | 0.8211 | 0.9999 |
101
- | 2.6338 | 9.0349 | 4400 | 2.2440 | 0.7884 | 1.0 |
102
- | 2.5375 | 9.2402 | 4500 | 2.2422 | 0.7863 | 0.9998 |
103
- | 2.516 | 9.4456 | 4600 | 2.2282 | 0.7863 | 1.0 |
104
- | 2.5467 | 9.6509 | 4700 | 2.2263 | 0.7656 | 0.9998 |
105
- | 2.53 | 9.8563 | 4800 | 2.2203 | 0.7556 | 0.9997 |
106
 
107
 
108
  ### Framework versions
 
1
  ---
2
  library_name: transformers
 
 
3
  tags:
4
  - generated_from_trainer
5
  metrics:
 
14
 
15
  # ssc-aln-model
16
 
17
+ This model was trained from scratch on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 1.8647
20
+ - Cer: 0.5679
21
+ - Wer: 0.9768
22
 
23
  ## Model description
24
 
 
38
 
39
  The following hyperparameters were used during training:
40
  - learning_rate: 0.0003
41
+ - train_batch_size: 4
42
  - eval_batch_size: 8
43
  - seed: 42
44
  - gradient_accumulation_steps: 2
45
+ - total_train_batch_size: 8
46
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 100
49
+ - num_epochs: 5
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
55
  |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
56
+ | 2.6542 | 0.1027 | 100 | 2.7873 | 0.9127 | 1.0 |
57
+ | 2.7312 | 0.2054 | 200 | 2.6739 | 0.8102 | 0.9999 |
58
+ | 2.7166 | 0.3082 | 300 | 2.3973 | 0.8100 | 0.9998 |
59
+ | 2.6442 | 0.4109 | 400 | 2.5701 | 0.7773 | 0.9863 |
60
+ | 2.6282 | 0.5136 | 500 | 2.4967 | 0.7596 | 1.0 |
61
+ | 2.6793 | 0.6163 | 600 | 2.4465 | 0.8274 | 0.9988 |
62
+ | 2.5982 | 0.7191 | 700 | 2.4531 | 0.7070 | 0.9893 |
63
+ | 2.5929 | 0.8218 | 800 | 2.3973 | 0.7647 | 0.9991 |
64
+ | 2.6211 | 0.9245 | 900 | 2.3430 | 0.7394 | 0.9911 |
65
+ | 2.5614 | 1.0267 | 1000 | 2.2116 | 0.6708 | 0.9887 |
66
+ | 2.5421 | 1.1294 | 1100 | 2.1762 | 0.7062 | 0.9970 |
67
+ | 2.5272 | 1.2322 | 1200 | 2.1483 | 0.6747 | 0.9907 |
68
+ | 2.4457 | 1.3349 | 1300 | 2.1416 | 0.6783 | 0.9754 |
69
+ | 2.4582 | 1.4376 | 1400 | 2.1515 | 0.6323 | 0.9812 |
70
+ | 2.5182 | 1.5403 | 1500 | 2.1518 | 0.6933 | 0.9828 |
71
+ | 2.545 | 1.6430 | 1600 | 2.1046 | 0.6844 | 0.9948 |
72
+ | 2.4768 | 1.7458 | 1700 | 2.0930 | 0.6794 | 0.9971 |
73
+ | 2.437 | 1.8485 | 1800 | 2.0755 | 0.6974 | 0.9977 |
74
+ | 2.4652 | 1.9512 | 1900 | 2.0531 | 0.6387 | 0.9852 |
75
+ | 2.4666 | 2.0534 | 2000 | 2.0942 | 0.6326 | 0.9725 |
76
+ | 2.4098 | 2.1561 | 2100 | 2.1318 | 0.7399 | 0.9999 |
77
+ | 2.295 | 2.2589 | 2200 | 2.0930 | 0.6261 | 0.9975 |
78
+ | 2.3255 | 2.3616 | 2300 | 2.0553 | 0.6080 | 0.9830 |
79
+ | 2.3362 | 2.4643 | 2400 | 2.0664 | 0.6241 | 0.9820 |
80
+ | 2.324 | 2.5670 | 2500 | 2.0415 | 0.6090 | 0.9839 |
81
+ | 2.3254 | 2.6697 | 2600 | 2.0766 | 0.5845 | 0.9765 |
82
+ | 2.3232 | 2.7725 | 2700 | 2.0245 | 0.6318 | 0.9836 |
83
+ | 2.2821 | 2.8752 | 2800 | 1.9850 | 0.6249 | 0.9870 |
84
+ | 2.2661 | 2.9779 | 2900 | 1.9709 | 0.6247 | 0.9770 |
85
+ | 2.2066 | 3.0801 | 3000 | 2.0029 | 0.5864 | 0.9691 |
86
+ | 2.1706 | 3.1828 | 3100 | 1.9698 | 0.5725 | 0.9681 |
87
+ | 2.1382 | 3.2856 | 3200 | 1.9499 | 0.5990 | 0.9759 |
88
+ | 2.2142 | 3.3883 | 3300 | 1.9464 | 0.6189 | 0.9825 |
89
+ | 2.2512 | 3.4910 | 3400 | 1.9367 | 0.6020 | 0.9843 |
90
+ | 2.1671 | 3.5937 | 3500 | 1.9393 | 0.5939 | 0.9799 |
91
+ | 2.2047 | 3.6965 | 3600 | 1.9381 | 0.5728 | 0.9700 |
92
+ | 2.1303 | 3.7992 | 3700 | 1.9116 | 0.5683 | 0.9722 |
93
+ | 2.1517 | 3.9019 | 3800 | 1.9412 | 0.5383 | 0.9495 |
94
+ | 2.2205 | 4.0041 | 3900 | 1.8760 | 0.5827 | 0.9780 |
95
+ | 2.07 | 4.1068 | 4000 | 1.9216 | 0.5793 | 0.9768 |
96
+ | 2.049 | 4.2096 | 4100 | 1.9057 | 0.5595 | 0.9694 |
97
+ | 2.057 | 4.3123 | 4200 | 1.9335 | 0.5549 | 0.9664 |
98
+ | 2.0582 | 4.4150 | 4300 | 1.9117 | 0.5552 | 0.9675 |
99
+ | 2.0678 | 4.5177 | 4400 | 1.8778 | 0.5699 | 0.9767 |
100
+ | 2.0643 | 4.6204 | 4500 | 1.8775 | 0.5704 | 0.9778 |
101
+ | 1.9829 | 4.7232 | 4600 | 1.8712 | 0.5704 | 0.9780 |
102
+ | 2.0293 | 4.8259 | 4700 | 1.8655 | 0.5577 | 0.9695 |
103
+ | 2.0133 | 4.9286 | 4800 | 1.8647 | 0.5679 | 0.9768 |
104
 
105
 
106
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c955f24b6abf9d16ffd3742ab639d7e25232cf349c68708e2337aecf848258bc
3
  size 1261991980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38aae8460b9e9269ef20c3ee9e13fde8efff7a5204c0dddebd53ef5397cb4a26
3
  size 1261991980