Hugging Face's logo

MelanieKoe
/
w2v2-base-pretrained_lr1e-4_at0.8_da0.3

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
MelanieKoe commited on
Commit
e96341a
•
1 Parent(s): bec21d8

Model save

Browse files
Files changed (2) hide show
  1. README.md +20 -34
  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: 3.3227
21
- - Wer: 1.0
22
 
23
  ## Model description
24
 
@@ -43,42 +43,28 @@ 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: 100
47
- - num_epochs: 200
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
52
- | Training Loss | Epoch | Step | Validation Loss | Wer |
53
- |:-------------:|:------:|:----:|:---------------:|:---:|
54
- | 18.4732 | 7.14 | 100 | 3.5073 | 1.0 |
55
- | 3.2701 | 14.29 | 200 | 3.3516 | 1.0 |
56
- | 3.234 | 21.43 | 300 | 3.3874 | 1.0 |
57
- | 3.2196 | 28.57 | 400 | 3.3154 | 1.0 |
58
- | 3.2146 | 35.71 | 500 | 3.3741 | 1.0 |
59
- | 3.1916 | 42.86 | 600 | 3.3402 | 1.0 |
60
- | 3.1749 | 50.0 | 700 | 3.3768 | 1.0 |
61
- | 3.2124 | 57.14 | 800 | 3.2787 | 1.0 |
62
- | 3.2133 | 64.29 | 900 | 3.3424 | 1.0 |
63
- | 3.2101 | 71.43 | 1000 | 3.2885 | 1.0 |
64
- | 3.2047 | 78.57 | 1100 | 3.3397 | 1.0 |
65
- | 3.2137 | 85.71 | 1200 | 3.3420 | 1.0 |
66
- | 3.2107 | 92.86 | 1300 | 3.3670 | 1.0 |
67
- | 3.214 | 100.0 | 1400 | 3.3011 | 1.0 |
68
- | 3.2089 | 107.14 | 1500 | 3.3278 | 1.0 |
69
- | 3.2095 | 114.29 | 1600 | 3.3363 | 1.0 |
70
- | 3.2053 | 121.43 | 1700 | 3.3086 | 1.0 |
71
- | 3.2009 | 128.57 | 1800 | 3.3105 | 1.0 |
72
- | 3.2037 | 135.71 | 1900 | 3.3275 | 1.0 |
73
- | 3.2074 | 142.86 | 2000 | 3.3092 | 1.0 |
74
- | 3.2134 | 150.0 | 2100 | 3.3099 | 1.0 |
75
- | 3.2042 | 157.14 | 2200 | 3.3025 | 1.0 |
76
- | 3.2006 | 164.29 | 2300 | 3.3110 | 1.0 |
77
- | 3.2052 | 171.43 | 2400 | 3.3393 | 1.0 |
78
- | 3.2052 | 178.57 | 2500 | 3.3191 | 1.0 |
79
- | 3.2026 | 185.71 | 2600 | 3.3303 | 1.0 |
80
- | 3.2113 | 192.86 | 2700 | 3.3261 | 1.0 |
81
- | 3.2048 | 200.0 | 2800 | 3.3227 | 1.0 |
82
 
83
 
84
  ### 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: 2.2956
21
+ - Wer: 0.1811
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: 3500
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
53
+ |:-------------:|:------:|:----:|:---------------:|:------:|
54
+ | 21.7228 | 17.86 | 250 | 3.7139 | 1.0 |
55
+ | 3.1757 | 35.71 | 500 | 3.1149 | 1.0 |
56
+ | 2.8079 | 53.57 | 750 | 1.7093 | 1.0038 |
57
+ | 0.301 | 71.43 | 1000 | 1.4764 | 0.2443 |
58
+ | 0.0637 | 89.29 | 1250 | 1.7091 | 0.2200 |
59
+ | 0.0311 | 107.14 | 1500 | 1.7022 | 0.2055 |
60
+ | 0.021 | 125.0 | 1750 | 2.0876 | 0.2016 |
61
+ | 0.0154 | 142.86 | 2000 | 2.2718 | 0.1974 |
62
+ | 0.0114 | 160.71 | 2250 | 2.1541 | 0.1845 |
63
+ | 0.0089 | 178.57 | 2500 | 2.2868 | 0.1854 |
64
+ | 0.0074 | 196.43 | 2750 | 2.3831 | 0.1914 |
65
+ | 0.0062 | 214.29 | 3000 | 2.2381 | 0.1841 |
66
+ | 0.0054 | 232.14 | 3250 | 2.4147 | 0.1824 |
67
+ | 0.0056 | 250.0 | 3500 | 2.2956 | 0.1811 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
68
 
69
 
70
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:58eacb22e5a83d65670bb6e503349cc36ffab5842bac9d57bc29de529bd74b72
3
  size 377629528
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f2a1ec108b95c1e87d1a43e7c98391129d78a8e44ab675eefdbd3be89e4943b
3
  size 377629528