Hugging Face's logo

golesheed
/
whisper-non-native-children-2-dutch

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Dutch
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
golesheed commited on
Commit
98654f0
·
verified ·
1 Parent(s): 74c4cf0

End of training

Browse files
Files changed (3) hide show
  1. README.md +10 -10
  2. model-00001-of-00002.safetensors +1 -1
  3. model-00002-of-00002.safetensors +1 -1
README.md CHANGED
@@ -2,11 +2,11 @@
2
  language:
3
  - nl
4
  license: apache-2.0
 
5
  tags:
6
  - generated_from_trainer
7
  metrics:
8
  - wer
9
- base_model: openai/whisper-large-v2
10
  model-index:
11
  - name: Whisper Large V2
12
  results: []
@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.4194
23
- - Wer: 16.8706
24
 
25
  ## Model description
26
 
@@ -52,13 +52,13 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer |
54
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
55
- | 0.6698 | 0.71 | 30 | 0.4438 | 26.2514 |
56
- | 0.2818 | 1.43 | 60 | 0.3862 | 17.4082 |
57
- | 0.1736 | 2.14 | 90 | 0.3823 | 14.6274 |
58
- | 0.0846 | 2.86 | 120 | 0.3899 | 15.7768 |
59
- | 0.04 | 3.57 | 150 | 0.4078 | 18.5206 |
60
- | 0.025 | 4.29 | 180 | 0.4031 | 15.6285 |
61
- | 0.0115 | 5.0 | 210 | 0.4194 | 16.8706 |
62
 
63
 
64
  ### Framework versions
 
2
  language:
3
  - nl
4
  license: apache-2.0
5
+ base_model: openai/whisper-large-v2
6
  tags:
7
  - generated_from_trainer
8
  metrics:
9
  - wer
 
10
  model-index:
11
  - name: Whisper Large V2
12
  results: []
 
19
 
20
  This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.4076
23
+ - Wer: 12.3813
24
 
25
  ## Model description
26
 
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer |
54
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
55
+ | 0.6749 | 0.71 | 30 | 0.3798 | 17.3625 |
56
+ | 0.26 | 1.43 | 60 | 0.3843 | 14.0477 |
57
+ | 0.163 | 2.14 | 90 | 0.3617 | 12.5963 |
58
+ | 0.0743 | 2.86 | 120 | 0.3539 | 13.2234 |
59
+ | 0.0429 | 3.57 | 150 | 0.3883 | 14.4598 |
60
+ | 0.024 | 4.29 | 180 | 0.4002 | 14.1014 |
61
+ | 0.011 | 5.0 | 210 | 0.4076 | 12.3813 |
62
 
63
 
64
  ### Framework versions
model-00001-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:834a19737a153f7b3497ad4dcbd1b5ead85e4bce2ffc93c5e5f541747d3dd1b6
3
  size 4992706480
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a05ce91dfee906ac36c1924687cf7bcaa7f1569accbb5f6745a7d861ba8c9971
3
  size 4992706480
model-00002-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:00717e0c163c9740690f2ddf2b9174b178fc45187d2e9f1fa8dd6dcafa1d0e90
3
  size 1180663192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:672f8b375a7c0ced3b1dd082ea76a630e8e0c7c646cf8bc7e3d4576af2f0af44
3
  size 1180663192