Hugging Face's logo

dongim04
/
wav2vec-large-en

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
dongim04 commited on
Commit
0fee992
•
1 Parent(s): 776fad8

End of training

Browse files
Files changed (3) hide show
  1. README.md +36 -52
  2. model.safetensors +1 -1
  3. runs/Nov16_05-01-48_b5d69e901b97/events.out.tfevents.1731733704.b5d69e901b97.1585.1 +2 -2
README.md CHANGED
@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # wav2vec-large-en
18
 
19
- This model is a fine-tuned version of [facebook/wav2vec2-large](https://huggingface.co/facebook/wav2vec2-large) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 7.2821
22
  - Wer: 1.0
23
 
24
  ## Model description
@@ -38,66 +38,50 @@ More information needed
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
- - learning_rate: 0.0001
42
  - train_batch_size: 32
43
  - eval_batch_size: 8
44
  - seed: 42
45
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - num_epochs: 30
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
52
- | Training Loss | Epoch | Step | Validation Loss | Wer |
53
- |:-------------:|:-------:|:----:|:---------------:|:---:|
54
- | 6.0558 | 0.7042 | 100 | 7.1035 | 1.0 |
55
- | 5.3678 | 1.4085 | 200 | 7.4719 | 1.0 |
56
- | 5.2548 | 2.1127 | 300 | 7.3891 | 1.0 |
57
- | 5.1982 | 2.8169 | 400 | 7.2118 | 1.0 |
58
- | 5.1951 | 3.5211 | 500 | 7.0877 | 1.0 |
59
- | 5.1829 | 4.2254 | 600 | 7.2987 | 1.0 |
60
- | 5.1937 | 4.9296 | 700 | 7.3988 | 1.0 |
61
- | 5.1905 | 5.6338 | 800 | 7.3964 | 1.0 |
62
- | 5.2445 | 6.3380 | 900 | 7.2698 | 1.0 |
63
- | 5.139 | 7.0423 | 1000 | 7.1558 | 1.0 |
64
- | 5.1925 | 7.7465 | 1100 | 7.2421 | 1.0 |
65
- | 5.2231 | 8.4507 | 1200 | 7.2611 | 1.0 |
66
- | 5.2901 | 9.1549 | 1300 | 7.3504 | 1.0 |
67
- | 5.0849 | 9.8592 | 1400 | 7.3211 | 1.0 |
68
- | 5.3853 | 10.5634 | 1500 | 7.1273 | 1.0 |
69
- | 5.2707 | 11.2676 | 1600 | 7.2369 | 1.0 |
70
- | 5.1156 | 11.9718 | 1700 | 7.3677 | 1.0 |
71
- | 5.1589 | 12.6761 | 1800 | 7.3260 | 1.0 |
72
- | 5.1839 | 13.3803 | 1900 | 7.2956 | 1.0 |
73
- | 5.3138 | 14.0845 | 2000 | 7.1808 | 1.0 |
74
- | 5.2344 | 14.7887 | 2100 | 7.2265 | 1.0 |
75
- | 5.1787 | 15.4930 | 2200 | 7.3030 | 1.0 |
76
- | 5.1879 | 16.1972 | 2300 | 7.2975 | 1.0 |
77
- | 5.106 | 16.9014 | 2400 | 7.2839 | 1.0 |
78
- | 5.3462 | 17.6056 | 2500 | 7.2212 | 1.0 |
79
- | 5.0687 | 18.3099 | 2600 | 7.2478 | 1.0 |
80
- | 5.1134 | 19.0141 | 2700 | 7.2222 | 1.0 |
81
- | 5.1133 | 19.7183 | 2800 | 7.3043 | 1.0 |
82
- | 5.2053 | 20.4225 | 2900 | 7.3150 | 1.0 |
83
- | 5.205 | 21.1268 | 3000 | 7.2460 | 1.0 |
84
- | 5.2558 | 21.8310 | 3100 | 7.2312 | 1.0 |
85
- | 5.215 | 22.5352 | 3200 | 7.2725 | 1.0 |
86
- | 5.1499 | 23.2394 | 3300 | 7.2609 | 1.0 |
87
- | 5.152 | 23.9437 | 3400 | 7.3052 | 1.0 |
88
- | 5.2532 | 24.6479 | 3500 | 7.2958 | 1.0 |
89
- | 5.3097 | 25.3521 | 3600 | 7.2818 | 1.0 |
90
- | 5.1221 | 26.0563 | 3700 | 7.2260 | 1.0 |
91
- | 5.2023 | 26.7606 | 3800 | 7.2648 | 1.0 |
92
- | 5.1029 | 27.4648 | 3900 | 7.2708 | 1.0 |
93
- | 5.085 | 28.1690 | 4000 | 7.2965 | 1.0 |
94
- | 5.2273 | 28.8732 | 4100 | 7.2845 | 1.0 |
95
- | 5.2394 | 29.5775 | 4200 | 7.2821 | 1.0 |
96
 
97
 
98
  ### Framework versions
99
 
100
- - Transformers 4.44.2
101
- - Pytorch 2.5.0+cu121
102
- - Datasets 3.1.0
103
- - Tokenizers 0.19.1
 
16
 
17
  # wav2vec-large-en
18
 
19
+ This model is a fine-tuned version of [facebook/wav2vec2-large](https://huggingface.co/facebook/wav2vec2-large) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 2.9034
22
  - Wer: 1.0
23
 
24
  ## Model description
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - learning_rate: 5e-06
42
  - train_batch_size: 32
43
  - eval_batch_size: 8
44
  - seed: 42
45
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
  - num_epochs: 30
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
53
+ |:-------------:|:------:|:----:|:---------------:|:---:|
54
+ | No log | 0.3521 | 50 | 3.5027 | 1.0 |
55
+ | 5.2013 | 0.7042 | 100 | 2.9715 | 1.0 |
56
+ | 5.2013 | 1.0563 | 150 | 3.0431 | 1.0 |
57
+ | 3.1541 | 1.4085 | 200 | 2.8890 | 1.0 |
58
+ | 3.1541 | 1.7606 | 250 | 2.9611 | 1.0 |
59
+ | 2.8986 | 2.1127 | 300 | 2.8821 | 1.0 |
60
+ | 2.8986 | 2.4648 | 350 | 2.8843 | 1.0 |
61
+ | 2.9209 | 2.8169 | 400 | 2.9202 | 1.0 |
62
+ | 2.9209 | 3.1690 | 450 | 2.9015 | 1.0 |
63
+ | 2.9283 | 3.5211 | 500 | 2.8956 | 1.0 |
64
+ | 2.9283 | 3.8732 | 550 | 2.8913 | 1.0 |
65
+ | 2.8832 | 4.2254 | 600 | 2.9140 | 1.0 |
66
+ | 2.8832 | 4.5775 | 650 | 2.8749 | 1.0 |
67
+ | 2.8789 | 4.9296 | 700 | 2.8970 | 1.0 |
68
+ | 2.8789 | 5.2817 | 750 | 2.9359 | 1.0 |
69
+ | 2.8753 | 5.6338 | 800 | 2.8821 | 1.0 |
70
+ | 2.8753 | 5.9859 | 850 | 2.8868 | 1.0 |
71
+ | 2.8859 | 6.3380 | 900 | 2.9063 | 1.0 |
72
+ | 2.8859 | 6.6901 | 950 | 2.8784 | 1.0 |
73
+ | 2.8693 | 7.0423 | 1000 | 2.9513 | 1.0 |
74
+ | 2.8693 | 7.3944 | 1050 | 2.8936 | 1.0 |
75
+ | 2.8911 | 7.7465 | 1100 | 2.8821 | 1.0 |
76
+ | 2.8911 | 8.0986 | 1150 | 2.9227 | 1.0 |
77
+ | 2.8625 | 8.4507 | 1200 | 2.8949 | 1.0 |
78
+ | 2.8625 | 8.8028 | 1250 | 2.9313 | 1.0 |
79
+ | 2.8641 | 9.1549 | 1300 | 2.8978 | 1.0 |
80
+ | 2.8641 | 9.5070 | 1350 | 2.9034 | 1.0 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
81
 
82
 
83
  ### Framework versions
84
 
85
+ - Transformers 4.46.2
86
+ - Pytorch 2.5.1+cu121
87
+ - Tokenizers 0.20.3
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cf29315bf9e9e82d4743401e069b9ca4d5d100c5c1ab2c280b027407320a1a67
3
  size 1261897368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5258868b0c9589b4313b8fb853ad72161d05463993ef9f13390faf83d100b7a6
3
  size 1261897368
runs/Nov16_05-01-48_b5d69e901b97/events.out.tfevents.1731733704.b5d69e901b97.1585.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:488fa7716a5d0a92722059ec7d94820380503811d968d4ec68d05b8ecdd5f6e9
3
- size 14960
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fab3f151be7ea2c87c8f7d69839e5e11c69fb86337897cb2118e7cafe9765bb
3
+ size 18173