Hugging Face's logo

DewiBrynJones
/
wav2vec2-xlsr-53-ft-btb-ccv-cy

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
DewiBrynJones commited on
Commit
9569f1c
•
1 Parent(s): 2d4b602

Model save

Browse files
Files changed (1) hide show
  1. README.md +106 -38
README.md CHANGED
@@ -2,8 +2,6 @@
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
5
- - automatic-speech-recognition
6
- - DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
7
  - generated_from_trainer
8
  metrics:
9
  - wer
@@ -17,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
19
 
20
- This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN-WITH-CCV - DEFAULT dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.5355
23
- - Wer: 0.4186
24
 
25
  ## Model description
26
 
@@ -46,43 +44,113 @@ The following hyperparameters were used during training:
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 500
49
- - training_steps: 3000
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss | Wer |
55
- |:-------------:|:------:|:----:|:---------------:|:------:|
56
- | No log | 0.0194 | 100 | 3.5494 | 1.0 |
57
- | No log | 0.0387 | 200 | 3.0426 | 1.0 |
58
- | No log | 0.0581 | 300 | 2.8965 | 1.0 |
59
- | No log | 0.0774 | 400 | 1.8263 | 0.9829 |
60
- | 3.9715 | 0.0968 | 500 | 1.3860 | 0.8749 |
61
- | 3.9715 | 0.1161 | 600 | 1.3084 | 0.8153 |
62
- | 3.9715 | 0.1355 | 700 | 1.0550 | 0.7337 |
63
- | 3.9715 | 0.1549 | 800 | 1.0012 | 0.7190 |
64
- | 3.9715 | 0.1742 | 900 | 0.9137 | 0.6752 |
65
- | 1.0155 | 0.1936 | 1000 | 0.8486 | 0.6469 |
66
- | 1.0155 | 0.2129 | 1100 | 0.8535 | 0.6112 |
67
- | 1.0155 | 0.2323 | 1200 | 0.8350 | 0.6193 |
68
- | 1.0155 | 0.2516 | 1300 | 0.7681 | 0.5670 |
69
- | 1.0155 | 0.2710 | 1400 | 0.7377 | 0.5559 |
70
- | 0.7987 | 0.2904 | 1500 | 0.7130 | 0.5437 |
71
- | 0.7987 | 0.3097 | 1600 | 0.7040 | 0.5452 |
72
- | 0.7987 | 0.3291 | 1700 | 0.6729 | 0.5051 |
73
- | 0.7987 | 0.3484 | 1800 | 0.6646 | 0.5113 |
74
- | 0.7987 | 0.3678 | 1900 | 0.6531 | 0.4969 |
75
- | 0.6851 | 0.3871 | 2000 | 0.6414 | 0.5038 |
76
- | 0.6851 | 0.4065 | 2100 | 0.6109 | 0.4677 |
77
- | 0.6851 | 0.4259 | 2200 | 0.6035 | 0.4692 |
78
- | 0.6851 | 0.4452 | 2300 | 0.5802 | 0.4590 |
79
- | 0.6851 | 0.4646 | 2400 | 0.5720 | 0.4455 |
80
- | 0.5979 | 0.4839 | 2500 | 0.5695 | 0.4426 |
81
- | 0.5979 | 0.5033 | 2600 | 0.5557 | 0.4351 |
82
- | 0.5979 | 0.5226 | 2700 | 0.5499 | 0.4270 |
83
- | 0.5979 | 0.5420 | 2800 | 0.5451 | 0.4258 |
84
- | 0.5979 | 0.5614 | 2900 | 0.5383 | 0.4217 |
85
- | 0.5753 | 0.5807 | 3000 | 0.5355 | 0.4186 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
86
 
87
 
88
  ### Framework versions
 
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
 
 
5
  - generated_from_trainer
6
  metrics:
7
  - wer
 
15
 
16
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
17
 
18
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.4061
21
+ - Wer: 0.3193
22
 
23
  ## Model description
24
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
  - lr_scheduler_warmup_steps: 500
47
+ - training_steps: 10000
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
53
+ |:-------------:|:------:|:-----:|:---------------:|:------:|
54
+ | No log | 0.0194 | 100 | 3.5565 | 1.0 |
55
+ | No log | 0.0387 | 200 | 3.0302 | 1.0 |
56
+ | No log | 0.0581 | 300 | 2.9461 | 1.0 |
57
+ | No log | 0.0774 | 400 | 1.8143 | 0.9407 |
58
+ | 3.9521 | 0.0968 | 500 | 1.4196 | 0.8693 |
59
+ | 3.9521 | 0.1161 | 600 | 1.1669 | 0.8055 |
60
+ | 3.9521 | 0.1355 | 700 | 1.0757 | 0.7596 |
61
+ | 3.9521 | 0.1549 | 800 | 0.9945 | 0.7224 |
62
+ | 3.9521 | 0.1742 | 900 | 0.9381 | 0.6871 |
63
+ | 1.0266 | 0.1936 | 1000 | 0.8978 | 0.6615 |
64
+ | 1.0266 | 0.2129 | 1100 | 0.8771 | 0.6450 |
65
+ | 1.0266 | 0.2323 | 1200 | 0.8516 | 0.6404 |
66
+ | 1.0266 | 0.2516 | 1300 | 0.8274 | 0.6138 |
67
+ | 1.0266 | 0.2710 | 1400 | 0.7993 | 0.5970 |
68
+ | 0.8454 | 0.2904 | 1500 | 0.7769 | 0.5888 |
69
+ | 0.8454 | 0.3097 | 1600 | 0.7664 | 0.5998 |
70
+ | 0.8454 | 0.3291 | 1700 | 0.7401 | 0.5592 |
71
+ | 0.8454 | 0.3484 | 1800 | 0.7465 | 0.5650 |
72
+ | 0.8454 | 0.3678 | 1900 | 0.7253 | 0.5791 |
73
+ | 0.7537 | 0.3871 | 2000 | 0.7039 | 0.5344 |
74
+ | 0.7537 | 0.4065 | 2100 | 0.6932 | 0.5168 |
75
+ | 0.7537 | 0.4259 | 2200 | 0.6969 | 0.5364 |
76
+ | 0.7537 | 0.4452 | 2300 | 0.6781 | 0.5174 |
77
+ | 0.7537 | 0.4646 | 2400 | 0.6761 | 0.5050 |
78
+ | 0.681 | 0.4839 | 2500 | 0.6721 | 0.5287 |
79
+ | 0.681 | 0.5033 | 2600 | 0.6598 | 0.5195 |
80
+ | 0.681 | 0.5226 | 2700 | 0.6555 | 0.4976 |
81
+ | 0.681 | 0.5420 | 2800 | 0.6535 | 0.4994 |
82
+ | 0.681 | 0.5614 | 2900 | 0.6259 | 0.4819 |
83
+ | 0.6737 | 0.5807 | 3000 | 0.6299 | 0.4802 |
84
+ | 0.6737 | 0.6001 | 3100 | 0.6379 | 0.4893 |
85
+ | 0.6737 | 0.6194 | 3200 | 0.6226 | 0.4806 |
86
+ | 0.6737 | 0.6388 | 3300 | 0.6089 | 0.4627 |
87
+ | 0.6737 | 0.6581 | 3400 | 0.6029 | 0.4735 |
88
+ | 0.6419 | 0.6775 | 3500 | 0.5871 | 0.4592 |
89
+ | 0.6419 | 0.6969 | 3600 | 0.6001 | 0.4611 |
90
+ | 0.6419 | 0.7162 | 3700 | 0.5849 | 0.4473 |
91
+ | 0.6419 | 0.7356 | 3800 | 0.5924 | 0.4638 |
92
+ | 0.6419 | 0.7549 | 3900 | 0.5768 | 0.4585 |
93
+ | 0.6183 | 0.7743 | 4000 | 0.5673 | 0.4453 |
94
+ | 0.6183 | 0.7937 | 4100 | 0.5575 | 0.4452 |
95
+ | 0.6183 | 0.8130 | 4200 | 0.5632 | 0.4475 |
96
+ | 0.6183 | 0.8324 | 4300 | 0.5499 | 0.4401 |
97
+ | 0.6183 | 0.8517 | 4400 | 0.5663 | 0.4310 |
98
+ | 0.5877 | 0.8711 | 4500 | 0.5585 | 0.4317 |
99
+ | 0.5877 | 0.8904 | 4600 | 0.5464 | 0.4200 |
100
+ | 0.5877 | 0.9098 | 4700 | 0.5381 | 0.4192 |
101
+ | 0.5877 | 0.9292 | 4800 | 0.5454 | 0.4202 |
102
+ | 0.5877 | 0.9485 | 4900 | 0.5238 | 0.4124 |
103
+ | 0.5621 | 0.9679 | 5000 | 0.5304 | 0.4135 |
104
+ | 0.5621 | 0.9872 | 5100 | 0.5163 | 0.4061 |
105
+ | 0.5621 | 1.0066 | 5200 | 0.5160 | 0.3993 |
106
+ | 0.5621 | 1.0259 | 5300 | 0.5089 | 0.3899 |
107
+ | 0.5621 | 1.0453 | 5400 | 0.5111 | 0.3986 |
108
+ | 0.4882 | 1.0647 | 5500 | 0.5010 | 0.3857 |
109
+ | 0.4882 | 1.0840 | 5600 | 0.4941 | 0.3859 |
110
+ | 0.4882 | 1.1034 | 5700 | 0.4940 | 0.3813 |
111
+ | 0.4882 | 1.1227 | 5800 | 0.4914 | 0.3782 |
112
+ | 0.4882 | 1.1421 | 5900 | 0.4875 | 0.3745 |
113
+ | 0.4569 | 1.1614 | 6000 | 0.4842 | 0.3807 |
114
+ | 0.4569 | 1.1808 | 6100 | 0.4861 | 0.3737 |
115
+ | 0.4569 | 1.2002 | 6200 | 0.4814 | 0.3761 |
116
+ | 0.4569 | 1.2195 | 6300 | 0.4781 | 0.3741 |
117
+ | 0.4569 | 1.2389 | 6400 | 0.4771 | 0.3682 |
118
+ | 0.4416 | 1.2582 | 6500 | 0.4710 | 0.3734 |
119
+ | 0.4416 | 1.2776 | 6600 | 0.4721 | 0.3660 |
120
+ | 0.4416 | 1.2969 | 6700 | 0.4679 | 0.3639 |
121
+ | 0.4416 | 1.3163 | 6800 | 0.4623 | 0.3665 |
122
+ | 0.4416 | 1.3357 | 6900 | 0.4611 | 0.3602 |
123
+ | 0.4324 | 1.3550 | 7000 | 0.4689 | 0.3609 |
124
+ | 0.4324 | 1.3744 | 7100 | 0.4573 | 0.3603 |
125
+ | 0.4324 | 1.3937 | 7200 | 0.4575 | 0.3546 |
126
+ | 0.4324 | 1.4131 | 7300 | 0.4556 | 0.3584 |
127
+ | 0.4324 | 1.4324 | 7400 | 0.4496 | 0.3507 |
128
+ | 0.4255 | 1.4518 | 7500 | 0.4461 | 0.3467 |
129
+ | 0.4255 | 1.4712 | 7600 | 0.4434 | 0.3462 |
130
+ | 0.4255 | 1.4905 | 7700 | 0.4436 | 0.3516 |
131
+ | 0.4255 | 1.5099 | 7800 | 0.4406 | 0.3458 |
132
+ | 0.4255 | 1.5292 | 7900 | 0.4387 | 0.3439 |
133
+ | 0.4094 | 1.5486 | 8000 | 0.4325 | 0.3410 |
134
+ | 0.4094 | 1.5679 | 8100 | 0.4360 | 0.3419 |
135
+ | 0.4094 | 1.5873 | 8200 | 0.4286 | 0.3377 |
136
+ | 0.4094 | 1.6067 | 8300 | 0.4301 | 0.3336 |
137
+ | 0.4094 | 1.6260 | 8400 | 0.4297 | 0.3323 |
138
+ | 0.4018 | 1.6454 | 8500 | 0.4270 | 0.3339 |
139
+ | 0.4018 | 1.6647 | 8600 | 0.4267 | 0.3320 |
140
+ | 0.4018 | 1.6841 | 8700 | 0.4224 | 0.3328 |
141
+ | 0.4018 | 1.7034 | 8800 | 0.4207 | 0.3298 |
142
+ | 0.4018 | 1.7228 | 8900 | 0.4197 | 0.3298 |
143
+ | 0.3899 | 1.7422 | 9000 | 0.4184 | 0.3258 |
144
+ | 0.3899 | 1.7615 | 9100 | 0.4165 | 0.3262 |
145
+ | 0.3899 | 1.7809 | 9200 | 0.4118 | 0.3229 |
146
+ | 0.3899 | 1.8002 | 9300 | 0.4134 | 0.3232 |
147
+ | 0.3899 | 1.8196 | 9400 | 0.4127 | 0.3209 |
148
+ | 0.3665 | 1.8389 | 9500 | 0.4108 | 0.3211 |
149
+ | 0.3665 | 1.8583 | 9600 | 0.4090 | 0.3199 |
150
+ | 0.3665 | 1.8777 | 9700 | 0.4076 | 0.3209 |
151
+ | 0.3665 | 1.8970 | 9800 | 0.4065 | 0.3198 |
152
+ | 0.3665 | 1.9164 | 9900 | 0.4062 | 0.3192 |
153
+ | 0.3698 | 1.9357 | 10000 | 0.4061 | 0.3193 |
154
 
155
 
156
  ### Framework versions