Hugging Face's logo

ivillar
/
amh_finetune

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
ivillar commited on
Commit
21593d5
•
1 Parent(s): b421e9d

End of training

Browse files
Files changed (3) hide show
  1. README.md +56 -56
  2. model.safetensors +1 -1
  3. runs/May22_18-19-51_652f2498e5b0/events.out.tfevents.1716402051.652f2498e5b0.10363.0 +2 -2
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Wer
24
  type: wer
25
- value: 100.0
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the ml-superb-subset dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 3.7471
36
- - Wer: 100.0
37
 
38
  ## Model description
39
 
@@ -52,7 +52,7 @@ More information needed
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
- - learning_rate: 0.005
56
  - train_batch_size: 64
57
  - eval_batch_size: 8
58
  - seed: 42
@@ -66,58 +66,58 @@ The following hyperparameters were used during training:
66
 
67
  ### Training results
68
 
69
- | Training Loss | Epoch | Step | Validation Loss | Wer |
70
- |:-------------:|:--------:|:----:|:---------------:|:-----:|
71
- | 17.1482 | 2.2222 | 10 | 9.2733 | 100.0 |
72
- | 4.0207 | 4.4444 | 20 | 3.9380 | 100.0 |
73
- | 5.0108 | 6.6667 | 30 | 4.5177 | 100.0 |
74
- | 5.4723 | 8.8889 | 40 | 4.1051 | 100.0 |
75
- | 3.9167 | 11.1111 | 50 | 3.8586 | 100.0 |
76
- | 3.8376 | 13.3333 | 60 | 3.8304 | 100.0 |
77
- | 3.8317 | 15.5556 | 70 | 3.8322 | 100.0 |
78
- | 3.8352 | 17.7778 | 80 | 3.8165 | 100.0 |
79
- | 3.8337 | 20.0 | 90 | 3.8249 | 100.0 |
80
- | 3.8231 | 22.2222 | 100 | 3.8129 | 100.0 |
81
- | 3.8169 | 24.4444 | 110 | 3.8217 | 100.0 |
82
- | 3.8195 | 26.6667 | 120 | 3.8120 | 100.0 |
83
- | 3.8186 | 28.8889 | 130 | 3.8160 | 100.0 |
84
- | 3.8271 | 31.1111 | 140 | 3.8150 | 100.0 |
85
- | 3.8273 | 33.3333 | 150 | 3.8134 | 100.0 |
86
- | 3.8186 | 35.5556 | 160 | 3.8138 | 100.0 |
87
- | 3.816 | 37.7778 | 170 | 3.8132 | 100.0 |
88
- | 3.817 | 40.0 | 180 | 3.8187 | 100.0 |
89
- | 3.8196 | 42.2222 | 190 | 3.8115 | 100.0 |
90
- | 3.8157 | 44.4444 | 200 | 3.8131 | 100.0 |
91
- | 3.814 | 46.6667 | 210 | 3.8158 | 100.0 |
92
- | 3.8198 | 48.8889 | 220 | 3.8073 | 100.0 |
93
- | 3.8169 | 51.1111 | 230 | 3.8036 | 100.0 |
94
- | 3.7913 | 53.3333 | 240 | 3.7904 | 100.0 |
95
- | 3.791 | 55.5556 | 250 | 3.7796 | 100.0 |
96
- | 3.765 | 57.7778 | 260 | 3.7673 | 100.0 |
97
- | 3.768 | 60.0 | 270 | 3.7672 | 100.0 |
98
- | 3.7594 | 62.2222 | 280 | 3.7594 | 100.0 |
99
- | 3.7488 | 64.4444 | 290 | 3.7601 | 100.0 |
100
- | 3.7506 | 66.6667 | 300 | 3.7583 | 100.0 |
101
- | 3.7408 | 68.8889 | 310 | 3.7580 | 100.0 |
102
- | 3.749 | 71.1111 | 320 | 3.7630 | 100.0 |
103
- | 3.7575 | 73.3333 | 330 | 3.7568 | 100.0 |
104
- | 3.7517 | 75.5556 | 340 | 3.7555 | 100.0 |
105
- | 3.758 | 77.7778 | 350 | 3.7536 | 100.0 |
106
- | 3.7491 | 80.0 | 360 | 3.7546 | 100.0 |
107
- | 3.7288 | 82.2222 | 370 | 3.7563 | 100.0 |
108
- | 3.7321 | 84.4444 | 380 | 3.7519 | 100.0 |
109
- | 3.7326 | 86.6667 | 390 | 3.7516 | 100.0 |
110
- | 3.7373 | 88.8889 | 400 | 3.7502 | 100.0 |
111
- | 3.7348 | 91.1111 | 410 | 3.7494 | 100.0 |
112
- | 3.7312 | 93.3333 | 420 | 3.7489 | 100.0 |
113
- | 3.732 | 95.5556 | 430 | 3.7477 | 100.0 |
114
- | 3.7431 | 97.7778 | 440 | 3.7489 | 100.0 |
115
- | 3.737 | 100.0 | 450 | 3.7474 | 100.0 |
116
- | 3.7364 | 102.2222 | 460 | 3.7479 | 100.0 |
117
- | 3.7265 | 104.4444 | 470 | 3.7478 | 100.0 |
118
- | 3.7339 | 106.6667 | 480 | 3.7471 | 100.0 |
119
- | 3.731 | 108.8889 | 490 | 3.7470 | 100.0 |
120
- | 3.736 | 111.1111 | 500 | 3.7471 | 100.0 |
121
 
122
 
123
  ### Framework versions
 
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 97.41641337386018
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the ml-superb-subset dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 2.8917
36
+ - Wer: 97.4164
37
 
38
  ## Model description
39
 
 
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
+ - learning_rate: 0.001
56
  - train_batch_size: 64
57
  - eval_batch_size: 8
58
  - seed: 42
 
66
 
67
  ### Training results
68
 
69
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
70
+ |:-------------:|:--------:|:----:|:---------------:|:--------:|
71
+ | 22.5796 | 2.2222 | 10 | 17.1583 | 100.0 |
72
+ | 9.5568 | 4.4444 | 20 | 7.4797 | 100.0 |
73
+ | 4.3875 | 6.6667 | 30 | 3.9841 | 100.0 |
74
+ | 3.8631 | 8.8889 | 40 | 3.8281 | 100.0 |
75
+ | 3.8298 | 11.1111 | 50 | 3.8117 | 100.0 |
76
+ | 3.7925 | 13.3333 | 60 | 3.7866 | 100.0 |
77
+ | 3.802 | 15.5556 | 70 | 3.7763 | 100.0 |
78
+ | 3.7845 | 17.7778 | 80 | 3.7681 | 100.0 |
79
+ | 3.7732 | 20.0 | 90 | 3.7627 | 100.0 |
80
+ | 3.7547 | 22.2222 | 100 | 3.7625 | 100.0 |
81
+ | 3.7471 | 24.4444 | 110 | 3.7588 | 100.0 |
82
+ | 3.7378 | 26.6667 | 120 | 3.7244 | 100.0 |
83
+ | 3.7278 | 28.8889 | 130 | 3.7337 | 100.0 |
84
+ | 3.71 | 31.1111 | 140 | 3.7188 | 100.0 |
85
+ | 3.6966 | 33.3333 | 150 | 3.7076 | 100.0 |
86
+ | 3.6811 | 35.5556 | 160 | 3.6916 | 100.0 |
87
+ | 3.6741 | 37.7778 | 170 | 3.6898 | 100.0 |
88
+ | 3.6337 | 40.0 | 180 | 3.6486 | 100.0 |
89
+ | 3.5766 | 42.2222 | 190 | 3.5913 | 100.0 |
90
+ | 3.5251 | 44.4444 | 200 | 3.5318 | 100.0 |
91
+ | 3.4533 | 46.6667 | 210 | 3.4549 | 100.0 |
92
+ | 3.3664 | 48.8889 | 220 | 3.3877 | 100.0 |
93
+ | 3.2963 | 51.1111 | 230 | 3.2852 | 100.0 |
94
+ | 3.1237 | 53.3333 | 240 | 3.1187 | 100.0 |
95
+ | 2.9356 | 55.5556 | 250 | 2.9620 | 100.0 |
96
+ | 2.7107 | 57.7778 | 260 | 2.7665 | 100.0 |
97
+ | 2.477 | 60.0 | 270 | 2.5155 | 99.3921 |
98
+ | 2.1786 | 62.2222 | 280 | 2.2953 | 98.4043 |
99
+ | 1.897 | 64.4444 | 290 | 2.1781 | 97.5684 |
100
+ | 1.6863 | 66.6667 | 300 | 2.1825 | 97.5684 |
101
+ | 1.4954 | 68.8889 | 310 | 2.1240 | 96.2766 |
102
+ | 1.3132 | 71.1111 | 320 | 2.1476 | 94.3769 |
103
+ | 1.1333 | 73.3333 | 330 | 2.2088 | 95.6687 |
104
+ | 0.9827 | 75.5556 | 340 | 2.2591 | 94.9088 |
105
+ | 0.9019 | 77.7778 | 350 | 2.4481 | 101.0638 |
106
+ | 0.7936 | 80.0 | 360 | 2.5467 | 103.4195 |
107
+ | 0.7015 | 82.2222 | 370 | 2.5279 | 95.5927 |
108
+ | 0.631 | 84.4444 | 380 | 2.6338 | 95.8207 |
109
+ | 0.5849 | 86.6667 | 390 | 2.6840 | 96.8085 |
110
+ | 0.5549 | 88.8889 | 400 | 2.7048 | 97.4164 |
111
+ | 0.5137 | 91.1111 | 410 | 2.7910 | 96.0486 |
112
+ | 0.4905 | 93.3333 | 420 | 2.8070 | 98.7842 |
113
+ | 0.4603 | 95.5556 | 430 | 2.8552 | 95.2888 |
114
+ | 0.457 | 97.7778 | 440 | 2.8382 | 95.8207 |
115
+ | 0.442 | 100.0 | 450 | 2.8831 | 98.2523 |
116
+ | 0.4437 | 102.2222 | 460 | 2.8800 | 97.5684 |
117
+ | 0.4346 | 104.4444 | 470 | 2.8805 | 97.7964 |
118
+ | 0.4341 | 106.6667 | 480 | 2.8864 | 97.6444 |
119
+ | 0.4319 | 108.8889 | 490 | 2.8911 | 97.3404 |
120
+ | 0.4403 | 111.1111 | 500 | 2.8917 | 97.4164 |
121
 
122
 
123
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:679341a418f55483f9c320eacd920895385782beba97e96b541cc397ce47bf9f
3
  size 1262758680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4c88bc93ba32c59972b6f237b3976568030b96fd4fcd77298af1098ccd6ea02
3
  size 1262758680
runs/May22_18-19-51_652f2498e5b0/events.out.tfevents.1716402051.652f2498e5b0.10363.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:26fa6209a3d4ad224751ea48dd8ae084d68363ec1ea32808e716bdaf02967052
3
- size 42338
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1becddb69471ee4130661d8d5bfc6c6743d9724f11a66d084870d5ba8df42dae
3
+ size 43432