Hugging Face's logo

MelanieKoe
/
w2v2-base-pretrained_lr5e-5_at0.2_da1

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
MelanieKoe commited on
Commit
9257fce
•
1 Parent(s): ca94e9e

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: 1.0942
21
- - Wer: 0.1674
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
- | 17.4656 | 3.91 | 250 | 3.8210 | 1.0 |
55
- | 3.2203 | 7.81 | 500 | 3.1655 | 1.0 |
56
- | 2.5403 | 11.72 | 750 | 1.2547 | 0.9979 |
57
- | 0.5746 | 15.62 | 1000 | 0.5996 | 0.5088 |
58
- | 0.2573 | 19.53 | 1250 | 0.7483 | 0.2046 |
59
- | 0.152 | 23.44 | 1500 | 0.9229 | 0.1862 |
60
- | 0.1082 | 27.34 | 1750 | 0.9192 | 0.1833 |
61
- | 0.0748 | 31.25 | 2000 | 1.0565 | 0.1747 |
62
- | 0.0603 | 35.16 | 2250 | 0.9710 | 0.1815 |
63
- | 0.0485 | 39.06 | 2500 | 1.0599 | 0.1704 |
64
- | 0.0399 | 42.97 | 2750 | 1.0942 | 0.1730 |
65
- | 0.034 | 46.88 | 3000 | 1.0842 | 0.1670 |
66
- | 0.0309 | 50.78 | 3250 | 1.0670 | 0.1632 |
67
- | 0.0269 | 54.69 | 3500 | 1.1369 | 0.1649 |
68
- | 0.0244 | 58.59 | 3750 | 1.0229 | 0.1666 |
69
- | 0.0228 | 62.5 | 4000 | 1.0942 | 0.1674 |
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: 1.3038
21
+ - Wer: 0.1709
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
+ | 19.5617 | 3.91 | 250 | 4.1984 | 1.0 |
55
+ | 3.3686 | 7.81 | 500 | 3.2319 | 1.0 |
56
+ | 3.1228 | 11.72 | 750 | 3.1341 | 1.0 |
57
+ | 2.9603 | 15.62 | 1000 | 2.3654 | 1.0 |
58
+ | 1.0738 | 19.53 | 1250 | 0.7578 | 0.5485 |
59
+ | 0.3549 | 23.44 | 1500 | 0.6579 | 0.2337 |
60
+ | 0.2219 | 27.34 | 1750 | 0.8304 | 0.1999 |
61
+ | 0.1579 | 31.25 | 2000 | 0.9428 | 0.1828 |
62
+ | 0.1216 | 35.16 | 2250 | 1.0046 | 0.1747 |
63
+ | 0.0958 | 39.06 | 2500 | 1.0114 | 0.1751 |
64
+ | 0.076 | 42.97 | 2750 | 1.2645 | 0.1768 |
65
+ | 0.0662 | 46.88 | 3000 | 1.2588 | 0.1739 |
66
+ | 0.0569 | 50.78 | 3250 | 1.3057 | 0.1730 |
67
+ | 0.0514 | 54.69 | 3500 | 1.2869 | 0.1696 |
68
+ | 0.0479 | 58.59 | 3750 | 1.2697 | 0.1704 |
69
+ | 0.0451 | 62.5 | 4000 | 1.3038 | 0.1709 |
70
 
71
 
72
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:683299d4bea66138d4775a6867752c1282a10a961c5857f18f135849c61df0bd
3
  size 377629528
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12f875cd558281541b1758e6f6d54ef33eaaf09464193a69a8e0c0b2d10c41fe
3
  size 377629528