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
196288f
•
1 Parent(s): e2876df

Model save

Browse files
Files changed (1) hide show
  1. README.md +107 -59
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: 0.4404
23
- - Wer: 0.3235
24
 
25
  ## Model description
26
 
@@ -45,64 +43,114 @@ The following hyperparameters were used during training:
45
  - seed: 42
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: 5000
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss | Wer |
55
- |:-------------:|:------:|:----:|:---------------:|:------:|
56
- | No log | 0.0566 | 100 | 3.5966 | 1.0 |
57
- | No log | 0.1133 | 200 | 3.5808 | 1.0 |
58
- | No log | 0.1699 | 300 | 2.2642 | 0.9884 |
59
- | No log | 0.2265 | 400 | 1.0231 | 0.7426 |
60
- | 3.6826 | 0.2831 | 500 | 0.9161 | 0.6457 |
61
- | 3.6826 | 0.3398 | 600 | 0.7942 | 0.5827 |
62
- | 3.6826 | 0.3964 | 700 | 0.7551 | 0.5417 |
63
- | 3.6826 | 0.4530 | 800 | 0.7243 | 0.5352 |
64
- | 3.6826 | 0.5096 | 900 | 0.6665 | 0.4985 |
65
- | 0.4714 | 0.5663 | 1000 | 0.6403 | 0.4863 |
66
- | 0.4714 | 0.6229 | 1100 | 0.6220 | 0.4727 |
67
- | 0.4714 | 0.6795 | 1200 | 0.6100 | 0.4642 |
68
- | 0.4714 | 0.7361 | 1300 | 0.5863 | 0.4442 |
69
- | 0.4714 | 0.7928 | 1400 | 0.5668 | 0.4384 |
70
- | 0.3661 | 0.8494 | 1500 | 0.5753 | 0.4421 |
71
- | 0.3661 | 0.9060 | 1600 | 0.5641 | 0.4311 |
72
- | 0.3661 | 0.9626 | 1700 | 0.5466 | 0.4153 |
73
- | 0.3661 | 1.0193 | 1800 | 0.5304 | 0.4023 |
74
- | 0.3661 | 1.0759 | 1900 | 0.5280 | 0.3938 |
75
- | 0.3069 | 1.1325 | 2000 | 0.5220 | 0.3902 |
76
- | 0.3069 | 1.1891 | 2100 | 0.5064 | 0.3929 |
77
- | 0.3069 | 1.2458 | 2200 | 0.5033 | 0.3823 |
78
- | 0.3069 | 1.3024 | 2300 | 0.4920 | 0.3773 |
79
- | 0.3069 | 1.3590 | 2400 | 0.4955 | 0.3737 |
80
- | 0.264 | 1.4156 | 2500 | 0.4955 | 0.3739 |
81
- | 0.264 | 1.4723 | 2600 | 0.4846 | 0.3651 |
82
- | 0.264 | 1.5289 | 2700 | 0.4761 | 0.3645 |
83
- | 0.264 | 1.5855 | 2800 | 0.4672 | 0.3627 |
84
- | 0.264 | 1.6421 | 2900 | 0.4645 | 0.3640 |
85
- | 0.2498 | 1.6988 | 3000 | 0.4652 | 0.3589 |
86
- | 0.2498 | 1.7554 | 3100 | 0.4596 | 0.3534 |
87
- | 0.2498 | 1.8120 | 3200 | 0.4505 | 0.3510 |
88
- | 0.2498 | 1.8686 | 3300 | 0.4490 | 0.3522 |
89
- | 0.2498 | 1.9253 | 3400 | 0.4460 | 0.3456 |
90
- | 0.2304 | 1.9819 | 3500 | 0.4440 | 0.3405 |
91
- | 0.2304 | 2.0385 | 3600 | 0.4493 | 0.3425 |
92
- | 0.2304 | 2.0951 | 3700 | 0.4386 | 0.3374 |
93
- | 0.2304 | 2.1518 | 3800 | 0.4424 | 0.3381 |
94
- | 0.2304 | 2.2084 | 3900 | 0.4385 | 0.3373 |
95
- | 0.1947 | 2.2650 | 4000 | 0.4394 | 0.3355 |
96
- | 0.1947 | 2.3216 | 4100 | 0.4354 | 0.3324 |
97
- | 0.1947 | 2.3783 | 4200 | 0.4345 | 0.3325 |
98
- | 0.1947 | 2.4349 | 4300 | 0.4343 | 0.3331 |
99
- | 0.1947 | 2.4915 | 4400 | 0.4517 | 0.3349 |
100
- | 0.1996 | 2.5481 | 4500 | 0.4571 | 0.3328 |
101
- | 0.1996 | 2.6048 | 4600 | 0.4423 | 0.3255 |
102
- | 0.1996 | 2.6614 | 4700 | 0.4405 | 0.3254 |
103
- | 0.1996 | 2.7180 | 4800 | 0.4437 | 0.3238 |
104
- | 0.1996 | 2.7746 | 4900 | 0.4401 | 0.3232 |
105
- | 0.2024 | 2.8313 | 5000 | 0.4404 | 0.3235 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
106
 
107
 
108
  ### 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: 2.8074
21
+ - Wer: 0.9983
22
 
23
  ## Model description
24
 
 
43
  - seed: 42
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 1000
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 | 0.0566 | 100 | 4.1582 | 1.0 |
55
+ | No log | 0.1133 | 200 | 3.1276 | 1.0 |
56
+ | No log | 0.1699 | 300 | 3.4072 | 1.0 |
57
+ | No log | 0.2265 | 400 | 2.1967 | 0.9899 |
58
+ | 4.6703 | 0.2831 | 500 | 1.0576 | 0.7174 |
59
+ | 4.6703 | 0.3398 | 600 | 0.8877 | 0.6405 |
60
+ | 4.6703 | 0.3964 | 700 | 0.7894 | 0.5861 |
61
+ | 4.6703 | 0.4530 | 800 | 0.7699 | 0.5877 |
62
+ | 4.6703 | 0.5096 | 900 | 0.7605 | 0.5403 |
63
+ | 0.5242 | 0.5663 | 1000 | 0.7429 | 0.5495 |
64
+ | 0.5242 | 0.6229 | 1100 | 0.6762 | 0.5081 |
65
+ | 0.5242 | 0.6795 | 1200 | 0.6703 | 0.5072 |
66
+ | 0.5242 | 0.7361 | 1300 | 0.6187 | 0.4623 |
67
+ | 0.5242 | 0.7928 | 1400 | 0.6205 | 0.4742 |
68
+ | 0.4093 | 0.8494 | 1500 | 0.6089 | 0.4607 |
69
+ | 0.4093 | 0.9060 | 1600 | 0.6079 | 0.4564 |
70
+ | 0.4093 | 0.9626 | 1700 | 0.5752 | 0.4487 |
71
+ | 0.4093 | 1.0193 | 1800 | 0.5519 | 0.4174 |
72
+ | 0.4093 | 1.0759 | 1900 | 0.5468 | 0.4107 |
73
+ | 0.3366 | 1.1325 | 2000 | 0.5372 | 0.4080 |
74
+ | 0.3366 | 1.1891 | 2100 | 0.5359 | 0.4072 |
75
+ | 0.3366 | 1.2458 | 2200 | 0.5304 | 0.4023 |
76
+ | 0.3366 | 1.3024 | 2300 | 0.5311 | 0.4011 |
77
+ | 0.3366 | 1.3590 | 2400 | 0.5186 | 0.3864 |
78
+ | 0.2939 | 1.4156 | 2500 | 0.5234 | 0.3934 |
79
+ | 0.2939 | 1.4723 | 2600 | 0.5213 | 0.3973 |
80
+ | 0.2939 | 1.5289 | 2700 | 0.5156 | 0.3877 |
81
+ | 0.2939 | 1.5855 | 2800 | 0.5052 | 0.3898 |
82
+ | 0.2939 | 1.6421 | 2900 | 0.4981 | 0.3833 |
83
+ | 0.2838 | 1.6988 | 3000 | 0.4990 | 0.3804 |
84
+ | 0.2838 | 1.7554 | 3100 | 0.5000 | 0.3807 |
85
+ | 0.2838 | 1.8120 | 3200 | 0.4961 | 0.3751 |
86
+ | 0.2838 | 1.8686 | 3300 | 0.4859 | 0.3731 |
87
+ | 0.2838 | 1.9253 | 3400 | 0.4812 | 0.3657 |
88
+ | 0.2694 | 1.9819 | 3500 | 0.4779 | 0.3620 |
89
+ | 0.2694 | 2.0385 | 3600 | 0.4943 | 0.3633 |
90
+ | 0.2694 | 2.0951 | 3700 | 0.4880 | 0.3677 |
91
+ | 0.2694 | 2.1518 | 3800 | 0.4990 | 0.3662 |
92
+ | 0.2694 | 2.2084 | 3900 | 0.5101 | 0.3699 |
93
+ | 0.2419 | 2.2650 | 4000 | 0.5393 | 0.3902 |
94
+ | 0.2419 | 2.3216 | 4100 | 0.6454 | 0.4513 |
95
+ | 0.2419 | 2.3783 | 4200 | 0.9892 | 0.5937 |
96
+ | 0.2419 | 2.4349 | 4300 | 0.7712 | 0.5167 |
97
+ | 0.2419 | 2.4915 | 4400 | 0.6338 | 0.4788 |
98
+ | 0.46 | 2.5481 | 4500 | 0.5562 | 0.4156 |
99
+ | 0.46 | 2.6048 | 4600 | 0.5377 | 0.3906 |
100
+ | 0.46 | 2.6614 | 4700 | 0.5687 | 0.4008 |
101
+ | 0.46 | 2.7180 | 4800 | 0.6321 | 0.4291 |
102
+ | 0.46 | 2.7746 | 4900 | 0.5834 | 0.4203 |
103
+ | 0.299 | 2.8313 | 5000 | 0.5302 | 0.3930 |
104
+ | 0.299 | 2.8879 | 5100 | 0.5316 | 0.3860 |
105
+ | 0.299 | 2.9445 | 5200 | 0.5344 | 0.3800 |
106
+ | 0.299 | 3.0011 | 5300 | 0.5349 | 0.3842 |
107
+ | 0.299 | 3.0578 | 5400 | 0.5776 | 0.4183 |
108
+ | 0.2839 | 3.1144 | 5500 | 0.5883 | 0.4100 |
109
+ | 0.2839 | 3.1710 | 5600 | 0.5723 | 0.4044 |
110
+ | 0.2839 | 3.2276 | 5700 | 0.5630 | 0.4078 |
111
+ | 0.2839 | 3.2843 | 5800 | 0.5810 | 0.4191 |
112
+ | 0.2839 | 3.3409 | 5900 | 0.5996 | 0.4228 |
113
+ | 0.3019 | 3.3975 | 6000 | 0.5682 | 0.4016 |
114
+ | 0.3019 | 3.4541 | 6100 | 0.5561 | 0.4057 |
115
+ | 0.3019 | 3.5108 | 6200 | 0.5905 | 0.4146 |
116
+ | 0.3019 | 3.5674 | 6300 | 0.5875 | 0.4190 |
117
+ | 0.3019 | 3.6240 | 6400 | 0.5878 | 0.4446 |
118
+ | 0.2944 | 3.6806 | 6500 | 0.5939 | 0.4404 |
119
+ | 0.2944 | 3.7373 | 6600 | 0.5903 | 0.4183 |
120
+ | 0.2944 | 3.7939 | 6700 | 0.5808 | 0.4059 |
121
+ | 0.2944 | 3.8505 | 6800 | 0.6155 | 0.4101 |
122
+ | 0.2944 | 3.9071 | 6900 | 0.7987 | 0.5823 |
123
+ | 0.3918 | 3.9638 | 7000 | 0.9750 | 0.5545 |
124
+ | 0.3918 | 4.0204 | 7100 | 1.0540 | 0.5689 |
125
+ | 0.3918 | 4.0770 | 7200 | 0.6851 | 0.4396 |
126
+ | 0.3918 | 4.1336 | 7300 | 0.7332 | 0.4973 |
127
+ | 0.3918 | 4.1903 | 7400 | 0.9466 | 0.6395 |
128
+ | 0.5378 | 4.2469 | 7500 | 0.8257 | 0.4851 |
129
+ | 0.5378 | 4.3035 | 7600 | 0.8490 | 0.4867 |
130
+ | 0.5378 | 4.3601 | 7700 | 0.8717 | 0.4711 |
131
+ | 0.5378 | 4.4168 | 7800 | 0.8839 | 0.5860 |
132
+ | 0.5378 | 4.4734 | 7900 | 2.9113 | 1.0 |
133
+ | 1.3847 | 4.5300 | 8000 | 2.8576 | 1.0 |
134
+ | 1.3847 | 4.5866 | 8100 | 2.8391 | 1.0 |
135
+ | 1.3847 | 4.6433 | 8200 | 2.8406 | 1.0 |
136
+ | 1.3847 | 4.6999 | 8300 | 2.8566 | 1.0 |
137
+ | 1.3847 | 4.7565 | 8400 | 2.8454 | 0.9998 |
138
+ | 2.8136 | 4.8131 | 8500 | 2.8340 | 0.9999 |
139
+ | 2.8136 | 4.8698 | 8600 | 2.8367 | 0.9999 |
140
+ | 2.8136 | 4.9264 | 8700 | 2.8334 | 0.9999 |
141
+ | 2.8136 | 4.9830 | 8800 | 2.8321 | 0.9999 |
142
+ | 2.8136 | 5.0396 | 8900 | 2.8025 | 0.9982 |
143
+ | 2.8007 | 5.0963 | 9000 | 2.8024 | 0.9975 |
144
+ | 2.8007 | 5.1529 | 9100 | 2.8043 | 0.9981 |
145
+ | 2.8007 | 5.2095 | 9200 | 2.8106 | 0.9992 |
146
+ | 2.8007 | 5.2661 | 9300 | 2.8067 | 0.9993 |
147
+ | 2.8007 | 5.3228 | 9400 | 2.8053 | 0.9986 |
148
+ | 2.7935 | 5.3794 | 9500 | 2.8077 | 0.9978 |
149
+ | 2.7935 | 5.4360 | 9600 | 2.8083 | 0.9987 |
150
+ | 2.7935 | 5.4926 | 9700 | 2.8080 | 0.9989 |
151
+ | 2.7935 | 5.5493 | 9800 | 2.8086 | 0.9986 |
152
+ | 2.7935 | 5.6059 | 9900 | 2.8079 | 0.9982 |
153
+ | 2.7861 | 5.6625 | 10000 | 2.8074 | 0.9983 |
154
 
155
 
156
  ### Framework versions