Hugging Face's logo

MelanieKoe
/
w2v2-base-pretrained_lr5e-5_at0.8_da0.4

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
MelanieKoe commited on
Commit
870995e
•
1 Parent(s): 2873321

Model save

Browse files
Files changed (2) hide show
  1. README.md +19 -19
  2. model.safetensors +1 -1
README.md CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 2.0085
21
- - Wer: 0.1892
22
 
23
  ## Model description
24
 
@@ -43,7 +43,7 @@ The following hyperparameters were used during training:
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: 500
47
  - training_steps: 4000
48
  - mixed_precision_training: Native AMP
49
 
@@ -51,22 +51,22 @@ The following hyperparameters were used during training:
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Wer |
53
  |:-------------:|:------:|:----:|:---------------:|:------:|
54
- | 17.5459 | 13.16 | 250 | 3.5740 | 1.0 |
55
- | 3.1637 | 26.32 | 500 | 3.2268 | 1.0 |
56
- | 2.4912 | 39.47 | 750 | 1.4668 | 0.8590 |
57
- | 0.2685 | 52.63 | 1000 | 1.2848 | 0.2469 |
58
- | 0.0883 | 65.79 | 1250 | 1.4088 | 0.2302 |
59
- | 0.0559 | 78.95 | 1500 | 1.5663 | 0.2196 |
60
- | 0.0386 | 92.11 | 1750 | 1.7690 | 0.2063 |
61
- | 0.0289 | 105.26 | 2000 | 1.7150 | 0.2114 |
62
- | 0.0244 | 118.42 | 2250 | 1.8159 | 0.2106 |
63
- | 0.018 | 131.58 | 2500 | 1.8855 | 0.2038 |
64
- | 0.0147 | 144.74 | 2750 | 1.8758 | 0.2042 |
65
- | 0.0121 | 157.89 | 3000 | 2.1326 | 0.2012 |
66
- | 0.0105 | 171.05 | 3250 | 1.9975 | 0.1897 |
67
- | 0.0093 | 184.21 | 3500 | 1.9094 | 0.1905 |
68
- | 0.0079 | 197.37 | 3750 | 2.0668 | 0.1909 |
69
- | 0.0082 | 210.53 | 4000 | 2.0085 | 0.1892 |
70
 
71
 
72
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 5.6336
21
+ - Wer: 0.8377
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: 4000
48
  - mixed_precision_training: Native AMP
49
 
 
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Wer |
53
  |:-------------:|:------:|:----:|:---------------:|:------:|
54
+ | 27.1555 | 13.16 | 250 | 16.9359 | 1.0 |
55
+ | 8.9276 | 26.32 | 500 | 9.0668 | 1.0 |
56
+ | 6.6067 | 39.47 | 750 | 8.0040 | 1.0 |
57
+ | 3.8775 | 52.63 | 1000 | 5.5239 | 1.0 |
58
+ | 3.1858 | 65.79 | 1250 | 5.5803 | 1.0 |
59
+ | 3.0238 | 78.95 | 1500 | 5.5821 | 1.0004 |
60
+ | 2.7704 | 92.11 | 1750 | 5.6283 | 1.0068 |
61
+ | 2.4867 | 105.26 | 2000 | 5.4805 | 0.9855 |
62
+ | 2.2251 | 118.42 | 2250 | 5.6338 | 0.9645 |
63
+ | 1.9731 | 131.58 | 2500 | 5.5967 | 0.9197 |
64
+ | 1.752 | 144.74 | 2750 | 5.5503 | 0.8949 |
65
+ | 1.5403 | 157.89 | 3000 | 5.4723 | 0.8663 |
66
+ | 1.3664 | 171.05 | 3250 | 5.5707 | 0.8603 |
67
+ | 1.2337 | 184.21 | 3500 | 5.5728 | 0.8449 |
68
+ | 1.1468 | 197.37 | 3750 | 5.6135 | 0.8402 |
69
+ | 1.0931 | 210.53 | 4000 | 5.6336 | 0.8377 |
70
 
71
 
72
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7750f579a043a3ab9b449029ccd91aa2e91e732b544db86cb3bd301dd5d6e716
3
  size 377629528
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0aa4355178b6842d277d03b631906928aabc046257333f6a329be5e9565abdc
3
  size 377629528