Hugging Face's logo

jun-han
/
Whisper-Small-architecture-change

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Chinese
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
jun-han commited on
Commit
ae64b89
•
1 Parent(s): e37be58

Model save

Browse files
Files changed (3) hide show
  1. README.md +41 -35
  2. model.safetensors +1 -1
  3. runs/Sep09_10-14-44_DESKTOP-Q3M97DH/events.out.tfevents.1725848091.DESKTOP-Q3M97DH.7208.0 +2 -2
README.md CHANGED
@@ -1,25 +1,23 @@
1
  ---
2
  library_name: transformers
3
- language:
4
- - zh
5
  license: apache-2.0
6
- base_model: openai/whisper-small
7
  tags:
8
  - generated_from_trainer
9
  model-index:
10
- - name: Whisper-Small-squeezeformer-architecture
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-squeezeformer-architecture
18
 
19
- This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Voice_Data_Collection_second_edition dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.4840
22
- - Cer: 27.2870
23
 
24
  ## Model description
25
 
@@ -45,37 +43,45 @@ The following hyperparameters were used during training:
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_steps: 3750
48
- - training_steps: 90000
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
- | Training Loss | Epoch | Step | Cer | Validation Loss |
54
- |:-------------:|:-----:|:-----:|:--------:|:---------------:|
55
- | 2.9532 | 1.0 | 3750 | 103.2182 | 2.9601 |
56
- | 1.6561 | 2.0 | 7500 | 85.2058 | 1.6430 |
57
- | 0.6747 | 3.0 | 11250 | 43.9073 | 0.7233 |
58
- | 0.4521 | 4.0 | 15000 | 33.8818 | 0.5573 |
59
- | 0.3412 | 5.0 | 18750 | 29.7393 | 0.4957 |
60
- | 0.2109 | 6.0 | 22500 | 27.6988 | 0.4640 |
61
- | 0.1365 | 7.0 | 26250 | 27.5348 | 0.4580 |
62
- | 0.105 | 8.0 | 30000 | 27.2348 | 0.4571 |
63
- | 0.4959 | 9.0 | 33750 | 24.2346 | 0.4091 |
64
- | 0.344 | 10.0 | 37500 | 22.3133 | 0.3801 |
65
- | 0.2431 | 11.0 | 41250 | 21.3667 | 0.3668 |
66
- | 0.1569 | 12.0 | 45000 | 21.1207 | 0.3665 |
67
- | 0.112 | 13.0 | 48750 | 21.1170 | 0.3702 |
68
- | 0.0716 | 14.0 | 52500 | 21.1263 | 0.3761 |
69
- | 0.052 | 15.0 | 56250 | 21.1822 | 0.3802 |
70
- | 0.038 | 16.0 | 60000 | 21.0778 | 0.3833 |
71
- | 0.2982 | 17.0 | 63750 | 24.5216 | 0.4189 |
72
- | 0.1896 | 18.0 | 67500 | 24.6688 | 0.4229 |
73
- | 0.155 | 19.0 | 71250 | 25.9154 | 0.4375 |
74
- | 0.1105 | 20.0 | 75000 | 26.1372 | 0.4476 |
75
- | 0.0727 | 21.0 | 78750 | 26.9087 | 0.4637 |
76
- | 0.0511 | 22.0 | 82500 | 0.4706 | 26.7894 |
77
- | 0.033 | 23.0 | 86250 | 0.4808 | 27.2180 |
78
- | 0.0246 | 24.0 | 90000 | 0.4840 | 27.2870 |
 
 
 
 
 
 
 
 
79
 
80
 
81
  ### Framework versions
 
1
  ---
2
  library_name: transformers
 
 
3
  license: apache-2.0
4
+ base_model: jun-han/Whisper-Small-architecture-change
5
  tags:
6
  - generated_from_trainer
7
  model-index:
8
+ - name: Whisper-Small-architecture-change
9
  results: []
10
  ---
11
 
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
  should probably proofread and complete it, then remove this comment. -->
14
 
15
+ # Whisper-Small-architecture-change
16
 
17
+ This model is a fine-tuned version of [jun-han/Whisper-Small-architecture-change](https://huggingface.co/jun-han/Whisper-Small-architecture-change) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.5061
20
+ - Cer: 28.1162
21
 
22
  ## Model description
23
 
 
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_steps: 3750
46
+ - training_steps: 120000
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
+ | Training Loss | Epoch | Step | Cer | Validation Loss |
52
+ |:-------------:|:-----:|:------:|:--------:|:---------------:|
53
+ | 2.9532 | 1.0 | 3750 | 103.2182 | 2.9601 |
54
+ | 1.6561 | 2.0 | 7500 | 85.2058 | 1.6430 |
55
+ | 0.6747 | 3.0 | 11250 | 43.9073 | 0.7233 |
56
+ | 0.4521 | 4.0 | 15000 | 33.8818 | 0.5573 |
57
+ | 0.3412 | 5.0 | 18750 | 29.7393 | 0.4957 |
58
+ | 0.2109 | 6.0 | 22500 | 27.6988 | 0.4640 |
59
+ | 0.1365 | 7.0 | 26250 | 27.5348 | 0.4580 |
60
+ | 0.105 | 8.0 | 30000 | 27.2348 | 0.4571 |
61
+ | 0.4959 | 9.0 | 33750 | 24.2346 | 0.4091 |
62
+ | 0.344 | 10.0 | 37500 | 22.3133 | 0.3801 |
63
+ | 0.2431 | 11.0 | 41250 | 21.3667 | 0.3668 |
64
+ | 0.1569 | 12.0 | 45000 | 21.1207 | 0.3665 |
65
+ | 0.112 | 13.0 | 48750 | 21.1170 | 0.3702 |
66
+ | 0.0716 | 14.0 | 52500 | 21.1263 | 0.3761 |
67
+ | 0.052 | 15.0 | 56250 | 21.1822 | 0.3802 |
68
+ | 0.038 | 16.0 | 60000 | 21.0778 | 0.3833 |
69
+ | 0.2982 | 17.0 | 63750 | 24.5216 | 0.4189 |
70
+ | 0.1896 | 18.0 | 67500 | 24.6688 | 0.4229 |
71
+ | 0.155 | 19.0 | 71250 | 25.9154 | 0.4375 |
72
+ | 0.1105 | 20.0 | 75000 | 26.1372 | 0.4476 |
73
+ | 0.0727 | 21.0 | 78750 | 26.9087 | 0.4637 |
74
+ | 0.0511 | 22.0 | 82500 | 26.7894 | 0.4706 |
75
+ | 0.033 | 23.0 | 86250 | 27.2180 | 0.4808 |
76
+ | 0.0246 | 24.0 | 90000 | 27.2870 | 0.4840 |
77
+ | 0.2775 | 25.0 | 93750 | 26.0310 | 0.4465 |
78
+ | 0.1631 | 26.0 | 97500 | 26.6068 | 0.4500 |
79
+ | 0.1428 | 27.0 | 101250 | 26.9869 | 0.4609 |
80
+ | 0.0955 | 28.0 | 105000 | 27.1919 | 0.4799 |
81
+ | 0.0756 | 29.0 | 108750 | 27.6261 | 0.4870 |
82
+ | 0.0584 | 30.0 | 112500 | 27.9634 | 0.4959 |
83
+ | 0.0386 | 31.0 | 116250 | 0.5041 | 28.1907 |
84
+ | 0.0367 | 32.0 | 120000 | 0.5061 | 28.1162 |
85
 
86
 
87
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:154aed22a527ba42f2263aaed0b2de0ce8f9248d23a61cbb66443956649dc43c
3
  size 1291395552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b012e10dfb339fe3433a746e221a51e9e590391017642e2414fd290a2c6f0b54
3
  size 1291395552
runs/Sep09_10-14-44_DESKTOP-Q3M97DH/events.out.tfevents.1725848091.DESKTOP-Q3M97DH.7208.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:535ecfcb6ceb01999bc094214613c1b22e2173e585a81e7620912d10b3bc0e21
3
- size 39332
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14287946002423656b2ff490a752ad03424c19b2868a9e64927b4a0c912c8ca9
3
+ size 72266