Hugging Face's logo

syp1229
/
whisper-small-ko-elder70

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
Hindi
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
syp1229 commited on
Commit
d74fc01
•
1 Parent(s): c73aa20

End of training

Browse files
Files changed (2) hide show
  1. README.md +23 -23
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -7,19 +7,19 @@ tags:
7
  - hf-asr-leaderboard
8
  - generated_from_trainer
9
  model-index:
10
- - name: Whisper Small ko-elder70 - syp1229
11
  results: []
12
  ---
13
 
14
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
  should probably proofread and complete it, then remove this comment. -->
16
 
17
- # Whisper Small ko-elder70 - syp1229
18
 
19
- This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the aihub elder over 70 dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.1961
22
- - Cer: 4.5511
23
 
24
  ## Model description
25
 
@@ -44,34 +44,34 @@ The following hyperparameters were used during training:
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
- - lr_scheduler_warmup_steps: 50
48
  - num_epochs: 5
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Cer |
53
  |:-------------:|:-----:|:----:|:---------------:|:------:|
54
- | 0.2246 | 0.3 | 100 | 0.2532 | 6.3965 |
55
- | 0.213 | 0.59 | 200 | 0.2291 | 6.2897 |
56
- | 0.1986 | 0.89 | 300 | 0.2037 | 5.5361 |
57
- | 0.0683 | 1.18 | 400 | 0.1912 | 4.7410 |
58
- | 0.0691 | 1.48 | 500 | 0.1967 | 5.0139 |
59
- | 0.0782 | 1.78 | 600 | 0.1999 | 5.0555 |
60
- | 0.0219 | 2.07 | 700 | 0.1907 | 4.9843 |
61
- | 0.0193 | 2.37 | 800 | 0.1918 | 4.5986 |
62
- | 0.0283 | 2.66 | 900 | 0.1966 | 4.8003 |
63
- | 0.0163 | 2.96 | 1000 | 0.1904 | 4.4740 |
64
- | 0.0083 | 3.25 | 1100 | 0.1958 | 4.3909 |
65
- | 0.0039 | 3.55 | 1200 | 0.1983 | 4.4562 |
66
- | 0.0043 | 3.85 | 1300 | 0.1944 | 4.4858 |
67
- | 0.0029 | 4.14 | 1400 | 0.1941 | 4.4384 |
68
- | 0.0027 | 4.44 | 1500 | 0.1981 | 4.5571 |
69
- | 0.003 | 4.73 | 1600 | 0.1961 | 4.5511 |
70
 
71
 
72
  ### Framework versions
73
 
74
  - Transformers 4.34.0.dev0
75
  - Pytorch 2.0.1+cu118
76
- - Datasets 2.14.4
77
  - Tokenizers 0.13.3
 
7
  - hf-asr-leaderboard
8
  - generated_from_trainer
9
  model-index:
10
+ - name: Whisper Small ko-Yfreq - syp1229
11
  results: []
12
  ---
13
 
14
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
  should probably proofread and complete it, then remove this comment. -->
16
 
17
+ # Whisper Small ko-Yfreq - syp1229
18
 
19
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the aihub Y dialogue dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.1951
22
+ - Cer: 0.0822
23
 
24
  ## Model description
25
 
 
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 100
48
  - num_epochs: 5
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Cer |
53
  |:-------------:|:-----:|:----:|:---------------:|:------:|
54
+ | 0.2218 | 0.3 | 100 | 0.2452 | 0.0622 |
55
+ | 0.2261 | 0.59 | 200 | 0.2389 | 0.0653 |
56
+ | 0.2032 | 0.89 | 300 | 0.2079 | 0.0554 |
57
+ | 0.0821 | 1.18 | 400 | 0.2152 | 0.0630 |
58
+ | 0.0763 | 1.48 | 500 | 0.2025 | 0.0520 |
59
+ | 0.0778 | 1.78 | 600 | 0.1997 | 0.0611 |
60
+ | 0.0216 | 2.07 | 700 | 0.1918 | 0.0609 |
61
+ | 0.0226 | 2.37 | 800 | 0.1884 | 0.0663 |
62
+ | 0.0289 | 2.66 | 900 | 0.1940 | 0.2490 |
63
+ | 0.0212 | 2.96 | 1000 | 0.1925 | 0.1068 |
64
+ | 0.0075 | 3.25 | 1100 | 0.1966 | 0.0615 |
65
+ | 0.0044 | 3.55 | 1200 | 0.1986 | 0.0629 |
66
+ | 0.0051 | 3.85 | 1300 | 0.1950 | 0.1097 |
67
+ | 0.003 | 4.14 | 1400 | 0.1929 | 0.0689 |
68
+ | 0.0029 | 4.44 | 1500 | 0.1958 | 0.0774 |
69
+ | 0.0029 | 4.73 | 1600 | 0.1951 | 0.0822 |
70
 
71
 
72
  ### Framework versions
73
 
74
  - Transformers 4.34.0.dev0
75
  - Pytorch 2.0.1+cu118
76
+ - Datasets 2.14.5
77
  - Tokenizers 0.13.3
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b0d6bd58eeec8da1eaeeff89eb287172cb1bc4ff7b272acfd46abd1521cb77c
3
  size 967102729
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86258819b4ec8be065b0de7cb102e0b6f4d89463c69944f377c952048ba4db5c
3
  size 967102729