Hugging Face's logo

TheRamsay
/
wav2vec2-gpt2-enc-dec

Automatic Speech Recognition
Transformers
Safetensors
speech-encoder-decoder
Generated from Trainer
Inference Endpoints
Model card Files Community
TheRamsay commited on
Commit
16ee4d9
·
verified ·
1 Parent(s): 74a461c

End of training

Browse files
Files changed (2) hide show
  1. README.md +80 -19
  2. generation_config.json +1 -0
README.md CHANGED
@@ -21,7 +21,7 @@ model-index:
21
  metrics:
22
  - name: Wer
23
  type: wer
24
- value: 0.6237000547345375
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
31
 
32
  This model is a fine-tuned version of [](https://huggingface.co/) on the common_voice_17_0 dataset.
33
  It achieves the following results on the evaluation set:
34
- - Loss: 0.2006
35
- - Wer: 0.6237
36
 
37
  ## Model description
38
 
@@ -51,12 +51,10 @@ More information needed
51
  ### Training hyperparameters
52
 
53
  The following hyperparameters were used during training:
54
- - learning_rate: 3e-05
55
- - train_batch_size: 16
56
- - eval_batch_size: 8
57
  - seed: 42
58
- - gradient_accumulation_steps: 2
59
- - total_train_batch_size: 32
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
  - lr_scheduler_warmup_ratio: 0.08
@@ -65,17 +63,80 @@ The following hyperparameters were used during training:
65
 
66
  ### Training results
67
 
68
- | Training Loss | Epoch | Step | Validation Loss | Wer |
69
- |:-------------:|:-------:|:-----:|:---------------:|:------:|
70
- | 0.3837 | 2.1942 | 2000 | 0.3241 | 0.8196 |
71
- | 0.3176 | 4.3884 | 4000 | 0.2855 | 0.7830 |
72
- | 0.2886 | 6.5826 | 6000 | 0.2620 | 0.7499 |
73
- | 0.2659 | 8.7767 | 8000 | 0.2431 | 0.7154 |
74
- | 0.2464 | 10.9709 | 10000 | 0.2285 | 0.6877 |
75
- | 0.2252 | 13.1651 | 12000 | 0.2163 | 0.6552 |
76
- | 0.2132 | 15.3593 | 14000 | 0.2087 | 0.6461 |
77
- | 0.2083 | 17.5535 | 16000 | 0.2032 | 0.6286 |
78
- | 0.2034 | 19.7477 | 18000 | 0.2006 | 0.6237 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
79
 
80
 
81
  ### Framework versions
 
21
  metrics:
22
  - name: Wer
23
  type: wer
24
+ value: 0.7337164750957854
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
 
32
  This model is a fine-tuned version of [](https://huggingface.co/) on the common_voice_17_0 dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: 0.3169
35
+ - Wer: 0.7337
36
 
37
  ## Model description
38
 
 
51
  ### Training hyperparameters
52
 
53
  The following hyperparameters were used during training:
54
+ - learning_rate: 1e-05
55
+ - train_batch_size: 4
56
+ - eval_batch_size: 4
57
  - seed: 42
 
 
58
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
  - lr_scheduler_type: linear
60
  - lr_scheduler_warmup_ratio: 0.08
 
63
 
64
  ### Training results
65
 
66
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
67
+ |:-------------:|:-------:|:------:|:---------------:|:------:|
68
+ | 0.7683 | 0.2744 | 2000 | 0.7136 | 1.0624 |
69
+ | 0.6998 | 0.5488 | 4000 | 0.6563 | 1.1959 |
70
+ | 0.6798 | 0.8232 | 6000 | 0.6385 | 1.2165 |
71
+ | 0.6563 | 1.0975 | 8000 | 0.6162 | 1.1502 |
72
+ | 0.6385 | 1.3719 | 10000 | 0.6097 | 1.2121 |
73
+ | 0.6509 | 1.6463 | 12000 | 0.6009 | 1.1585 |
74
+ | 0.6365 | 1.9207 | 14000 | 0.5816 | 1.1171 |
75
+ | 0.6148 | 2.1951 | 16000 | 0.5811 | 1.1667 |
76
+ | 0.6081 | 2.4695 | 18000 | 0.5620 | 1.0605 |
77
+ | 0.6116 | 2.7439 | 20000 | 0.5596 | 1.1314 |
78
+ | 0.5875 | 3.0182 | 22000 | 0.5496 | 1.1081 |
79
+ | 0.5831 | 3.2926 | 24000 | 0.5347 | 1.0435 |
80
+ | 0.5798 | 3.5670 | 26000 | 0.5363 | 1.0736 |
81
+ | 0.6004 | 3.8414 | 28000 | 0.5234 | 1.0120 |
82
+ | 0.5525 | 4.1158 | 30000 | 0.5096 | 0.9784 |
83
+ | 0.5616 | 4.3902 | 32000 | 0.5074 | 0.9910 |
84
+ | 0.5623 | 4.6646 | 34000 | 0.5054 | 0.9849 |
85
+ | 0.549 | 4.9389 | 36000 | 0.5028 | 0.9962 |
86
+ | 0.5334 | 5.2133 | 38000 | 0.4876 | 0.9537 |
87
+ | 0.5482 | 5.4877 | 40000 | 0.4865 | 0.9414 |
88
+ | 0.5277 | 5.7621 | 42000 | 0.4886 | 0.9789 |
89
+ | 0.5126 | 6.0365 | 44000 | 0.4769 | 0.9494 |
90
+ | 0.5246 | 6.3109 | 46000 | 0.4690 | 0.9278 |
91
+ | 0.5255 | 6.5853 | 48000 | 0.4688 | 0.9455 |
92
+ | 0.5156 | 6.8597 | 50000 | 0.4597 | 0.9236 |
93
+ | 0.5135 | 7.1340 | 52000 | 0.4581 | 0.9245 |
94
+ | 0.5041 | 7.4084 | 54000 | 0.4488 | 0.9067 |
95
+ | 0.505 | 7.6828 | 56000 | 0.4456 | 0.9050 |
96
+ | 0.5151 | 7.9572 | 58000 | 0.4441 | 0.9053 |
97
+ | 0.4953 | 8.2316 | 60000 | 0.4374 | 0.8990 |
98
+ | 0.4815 | 8.5060 | 62000 | 0.4338 | 0.8987 |
99
+ | 0.506 | 8.7804 | 64000 | 0.4295 | 0.8897 |
100
+ | 0.4847 | 9.0547 | 66000 | 0.4273 | 0.8922 |
101
+ | 0.4763 | 9.3291 | 68000 | 0.4206 | 0.8782 |
102
+ | 0.4822 | 9.6035 | 70000 | 0.4180 | 0.8790 |
103
+ | 0.4728 | 9.8779 | 72000 | 0.4131 | 0.8673 |
104
+ | 0.4645 | 10.1523 | 74000 | 0.4086 | 0.8708 |
105
+ | 0.4696 | 10.4267 | 76000 | 0.4087 | 0.8741 |
106
+ | 0.4664 | 10.7011 | 78000 | 0.4020 | 0.8547 |
107
+ | 0.4677 | 10.9754 | 80000 | 0.3974 | 0.8503 |
108
+ | 0.4482 | 11.2498 | 82000 | 0.3907 | 0.8410 |
109
+ | 0.4518 | 11.5242 | 84000 | 0.3890 | 0.8451 |
110
+ | 0.4638 | 11.7986 | 86000 | 0.3855 | 0.8407 |
111
+ | 0.4325 | 12.0730 | 88000 | 0.3825 | 0.8328 |
112
+ | 0.4421 | 12.3474 | 90000 | 0.3742 | 0.8134 |
113
+ | 0.4366 | 12.6218 | 92000 | 0.3751 | 0.8303 |
114
+ | 0.4412 | 12.8961 | 94000 | 0.3671 | 0.8120 |
115
+ | 0.4297 | 13.1705 | 96000 | 0.3639 | 0.8035 |
116
+ | 0.4267 | 13.4449 | 98000 | 0.3651 | 0.8131 |
117
+ | 0.437 | 13.7193 | 100000 | 0.3575 | 0.7947 |
118
+ | 0.4373 | 13.9937 | 102000 | 0.3561 | 0.8010 |
119
+ | 0.4182 | 14.2681 | 104000 | 0.3514 | 0.7860 |
120
+ | 0.4141 | 14.5425 | 106000 | 0.3490 | 0.7874 |
121
+ | 0.4117 | 14.8168 | 108000 | 0.3466 | 0.7833 |
122
+ | 0.4116 | 15.0912 | 110000 | 0.3446 | 0.7753 |
123
+ | 0.4074 | 15.3656 | 112000 | 0.3411 | 0.7696 |
124
+ | 0.4086 | 15.6400 | 114000 | 0.3394 | 0.7770 |
125
+ | 0.4189 | 15.9144 | 116000 | 0.3358 | 0.7666 |
126
+ | 0.3972 | 16.1888 | 118000 | 0.3332 | 0.7608 |
127
+ | 0.3911 | 16.4632 | 120000 | 0.3328 | 0.7603 |
128
+ | 0.4062 | 16.7375 | 122000 | 0.3296 | 0.7496 |
129
+ | 0.4 | 17.0119 | 124000 | 0.3279 | 0.7537 |
130
+ | 0.388 | 17.2863 | 126000 | 0.3262 | 0.7469 |
131
+ | 0.3929 | 17.5607 | 128000 | 0.3251 | 0.7471 |
132
+ | 0.4036 | 17.8351 | 130000 | 0.3237 | 0.7455 |
133
+ | 0.392 | 18.1095 | 132000 | 0.3217 | 0.7417 |
134
+ | 0.3916 | 18.3839 | 134000 | 0.3207 | 0.7406 |
135
+ | 0.3934 | 18.6583 | 136000 | 0.3190 | 0.7375 |
136
+ | 0.3805 | 18.9326 | 138000 | 0.3188 | 0.7384 |
137
+ | 0.3778 | 19.2070 | 140000 | 0.3178 | 0.7365 |
138
+ | 0.3795 | 19.4814 | 142000 | 0.3173 | 0.7337 |
139
+ | 0.3841 | 19.7558 | 144000 | 0.3169 | 0.7337 |
140
 
141
 
142
  ### Framework versions
generation_config.json CHANGED
@@ -1,6 +1,7 @@
1
  {
2
  "_from_model_config": true,
3
  "bos_token_id": 50256,
 
4
  "eos_token_id": 50256,
5
  "transformers_version": "4.45.2"
6
  }
 
1
  {
2
  "_from_model_config": true,
3
  "bos_token_id": 50256,
4
+ "decoder_start_token_id": 50256,
5
  "eos_token_id": 50256,
6
  "transformers_version": "4.45.2"
7
  }