Hugging Face's logo

MelanieKoe
/
w2v2-base-pretrained_lr5e-5_at0.4_da1

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
MelanieKoe commited on
Commit
bd064d3
•
1 Parent(s): debe51e

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: 1.4243
21
- - Wer: 0.2593
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
- | 17.3727 | 4.03 | 250 | 8.4434 | 1.0 |
55
- | 4.5158 | 8.06 | 500 | 5.6577 | 1.0 |
56
- | 3.3968 | 12.1 | 750 | 5.5711 | 1.0 |
57
- | 3.4087 | 16.13 | 1000 | 4.7248 | 1.0 |
58
- | 3.2781 | 20.16 | 1250 | 4.1466 | 1.0004 |
59
- | 3.0895 | 24.19 | 1500 | 4.0231 | 1.0004 |
60
- | 2.7559 | 28.23 | 1750 | 2.6579 | 0.9735 |
61
- | 1.8327 | 32.26 | 2000 | 1.5957 | 0.8979 |
62
- | 0.7991 | 36.29 | 2250 | 1.1338 | 0.5656 |
63
- | 0.4226 | 40.32 | 2500 | 1.1239 | 0.4088 |
64
- | 0.2746 | 44.35 | 2750 | 1.2772 | 0.3430 |
65
- | 0.1934 | 48.39 | 3000 | 1.2697 | 0.3187 |
66
- | 0.1437 | 52.42 | 3250 | 1.3526 | 0.3033 |
67
- | 0.1151 | 56.45 | 3500 | 1.3560 | 0.2777 |
68
- | 0.0975 | 60.48 | 3750 | 1.3470 | 0.2606 |
69
- | 0.0885 | 64.52 | 4000 | 1.4243 | 0.2593 |
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: 1.4372
21
+ - Wer: 0.1666
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
+ | 18.4042 | 4.03 | 250 | 4.2497 | 1.0 |
55
+ | 3.3741 | 8.06 | 500 | 3.2004 | 1.0 |
56
+ | 3.1004 | 12.1 | 750 | 3.1159 | 1.0 |
57
+ | 2.3298 | 16.13 | 1000 | 1.0486 | 0.7809 |
58
+ | 0.5044 | 20.16 | 1250 | 0.6083 | 0.3464 |
59
+ | 0.27 | 24.19 | 1500 | 0.6948 | 0.2456 |
60
+ | 0.1833 | 28.23 | 1750 | 0.9908 | 0.1956 |
61
+ | 0.1324 | 32.26 | 2000 | 1.0134 | 0.1995 |
62
+ | 0.1027 | 36.29 | 2250 | 1.3176 | 0.1760 |
63
+ | 0.0852 | 40.32 | 2500 | 1.1929 | 0.1837 |
64
+ | 0.0703 | 44.35 | 2750 | 1.3824 | 0.1670 |
65
+ | 0.0601 | 48.39 | 3000 | 1.3337 | 0.1674 |
66
+ | 0.0546 | 52.42 | 3250 | 1.3566 | 0.1717 |
67
+ | 0.05 | 56.45 | 3500 | 1.4653 | 0.1670 |
68
+ | 0.046 | 60.48 | 3750 | 1.4321 | 0.1696 |
69
+ | 0.0452 | 64.52 | 4000 | 1.4372 | 0.1666 |
70
 
71
 
72
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b0afc41370b4fddf42a27236233f4929efafd98674fa67f9d77ed737a08e80f8
3
  size 377629528
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eab7d3416c267c70b162ed50c62f4c42141635c80978c543d3ad82195a025a1e
3
  size 377629528