Hugging Face's logo

MelanieKoe
/
w2v2-base-pretrained_lr5e-5_at0.8_da0.7

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
MelanieKoe commited on
Commit
de3f6e1
•
1 Parent(s): 726331d

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: 1.7538
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
- | 16.4265 | 7.81 | 250 | 3.9246 | 1.0 |
55
- | 3.1637 | 15.62 | 500 | 3.1545 | 1.0 |
56
- | 2.2078 | 23.44 | 750 | 1.2345 | 0.6258 |
57
- | 0.2629 | 31.25 | 1000 | 1.0938 | 0.2862 |
58
- | 0.1013 | 39.06 | 1250 | 1.2497 | 0.2482 |
59
- | 0.061 | 46.88 | 1500 | 1.4009 | 0.1956 |
60
- | 0.0438 | 54.69 | 1750 | 1.4359 | 0.1935 |
61
- | 0.0334 | 62.5 | 2000 | 1.6005 | 0.2003 |
62
- | 0.0256 | 70.31 | 2250 | 1.5475 | 0.1897 |
63
- | 0.0231 | 78.12 | 2500 | 1.6901 | 0.1914 |
64
- | 0.0176 | 85.94 | 2750 | 1.6743 | 0.1961 |
65
- | 0.0164 | 93.75 | 3000 | 1.6498 | 0.1815 |
66
- | 0.0133 | 101.56 | 3250 | 1.7255 | 0.1897 |
67
- | 0.0127 | 109.38 | 3500 | 1.7270 | 0.1875 |
68
- | 0.011 | 117.19 | 3750 | 1.7833 | 0.1918 |
69
- | 0.0111 | 125.0 | 4000 | 1.7538 | 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: 2.2042
21
+ - Wer: 0.1884
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
+ | 24.3422 | 7.81 | 250 | 5.6070 | 1.0 |
55
+ | 3.5981 | 15.62 | 500 | 3.2535 | 1.0 |
56
+ | 3.1121 | 23.44 | 750 | 3.1577 | 1.0 |
57
+ | 3.0596 | 31.25 | 1000 | 3.1214 | 1.0 |
58
+ | 3.0143 | 39.06 | 1250 | 2.9603 | 1.0 |
59
+ | 1.4861 | 46.88 | 1500 | 1.2406 | 0.4007 |
60
+ | 0.2223 | 54.69 | 1750 | 1.3926 | 0.2324 |
61
+ | 0.1147 | 62.5 | 2000 | 1.5275 | 0.2136 |
62
+ | 0.0775 | 70.31 | 2250 | 1.8277 | 0.1986 |
63
+ | 0.0601 | 78.12 | 2500 | 1.9747 | 0.1944 |
64
+ | 0.0479 | 85.94 | 2750 | 2.0632 | 0.1909 |
65
+ | 0.042 | 93.75 | 3000 | 2.1333 | 0.1991 |
66
+ | 0.0353 | 101.56 | 3250 | 2.1743 | 0.1982 |
67
+ | 0.0315 | 109.38 | 3500 | 2.1585 | 0.1939 |
68
+ | 0.0274 | 117.19 | 3750 | 2.1521 | 0.1914 |
69
+ | 0.0279 | 125.0 | 4000 | 2.2042 | 0.1884 |
70
 
71
 
72
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9ffd9fc6e8f341bdbcf1974be1f4cf4869ed377247a88049ef446ba7d1043315
3
  size 377629528
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9be2b7f00ac5f0913585ccabd087e28d406c3a799f5eeafa4eb3e3b97747bb5c
3
  size 377629528