Hugging Face's logo

ruisp
/
whisper-tiny

Automatic Speech Recognition
Transformers
PyTorch
whisper
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
ruisp commited on
Commit
b011352
•
1 Parent(s): 5a8c02b

End of training

Browse files
Files changed (3) hide show
  1. README.md +17 -25
  2. pytorch_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Wer
24
  type: wer
25
- value: 0.564935064935065
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,9 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.7547
36
- - Wer Ortho: 0.5657
37
- - Wer: 0.5649
38
 
39
  ## Model description
40
 
@@ -60,32 +60,24 @@ The following hyperparameters were used during training:
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: constant_with_warmup
62
  - lr_scheduler_warmup_steps: 150
63
- - training_steps: 500
64
 
65
  ### Training results
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
68
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
69
- | 3.1392 | 0.89 | 25 | 2.1394 | 0.5392 | 0.4061 |
70
- | 1.1631 | 1.79 | 50 | 0.6562 | 0.5848 | 0.5567 |
71
- | 0.4252 | 2.68 | 75 | 0.5337 | 0.6496 | 0.6352 |
72
- | 0.2071 | 3.57 | 100 | 0.5282 | 0.4738 | 0.4616 |
73
- | 0.1267 | 4.46 | 125 | 0.5674 | 0.4300 | 0.4221 |
74
- | 0.0503 | 5.36 | 150 | 0.5919 | 0.3924 | 0.3819 |
75
- | 0.0336 | 6.25 | 175 | 0.6248 | 0.3689 | 0.3548 |
76
- | 0.0235 | 7.14 | 200 | 0.6359 | 0.3714 | 0.3654 |
77
- | 0.012 | 8.04 | 225 | 0.6404 | 0.3658 | 0.3548 |
78
- | 0.0152 | 8.93 | 250 | 0.7101 | 0.5515 | 0.5431 |
79
- | 0.0081 | 9.82 | 275 | 0.6697 | 0.3763 | 0.3660 |
80
- | 0.0098 | 10.71 | 300 | 0.6560 | 0.3541 | 0.3465 |
81
- | 0.0089 | 11.61 | 325 | 0.7286 | 0.5367 | 0.5419 |
82
- | 0.01 | 12.5 | 350 | 0.7133 | 0.4769 | 0.4764 |
83
- | 0.0086 | 13.39 | 375 | 0.7350 | 0.5435 | 0.5378 |
84
- | 0.0062 | 14.29 | 400 | 0.7670 | 0.4275 | 0.4286 |
85
- | 0.0039 | 15.18 | 425 | 0.7543 | 0.4553 | 0.4534 |
86
- | 0.0086 | 16.07 | 450 | 0.7541 | 0.4571 | 0.4551 |
87
- | 0.0043 | 16.96 | 475 | 0.7642 | 0.5527 | 0.5525 |
88
- | 0.0022 | 17.86 | 500 | 0.7547 | 0.5657 | 0.5649 |
89
 
90
 
91
  ### Framework versions
 
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 0.39315230224321135
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.7067
36
+ - Wer Ortho: 0.3991
37
+ - Wer: 0.3932
38
 
39
  ## Model description
40
 
 
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: constant_with_warmup
62
  - lr_scheduler_warmup_steps: 150
63
+ - training_steps: 300
64
 
65
  ### Training results
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
68
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
69
+ | 3.1391 | 0.89 | 25 | 2.1392 | 0.5392 | 0.4061 |
70
+ | 1.1631 | 1.79 | 50 | 0.6561 | 0.5811 | 0.5531 |
71
+ | 0.4253 | 2.68 | 75 | 0.5337 | 0.6397 | 0.6257 |
72
+ | 0.2072 | 3.57 | 100 | 0.5282 | 0.4756 | 0.4628 |
73
+ | 0.1265 | 4.46 | 125 | 0.5668 | 0.4300 | 0.4227 |
74
+ | 0.0507 | 5.36 | 150 | 0.5849 | 0.4485 | 0.4380 |
75
+ | 0.0357 | 6.25 | 175 | 0.6048 | 0.3584 | 0.3430 |
76
+ | 0.0243 | 7.14 | 200 | 0.6141 | 0.3590 | 0.3465 |
77
+ | 0.0154 | 8.04 | 225 | 0.6424 | 0.3701 | 0.3524 |
78
+ | 0.0189 | 8.93 | 250 | 0.6685 | 0.3732 | 0.3660 |
79
+ | 0.0154 | 9.82 | 275 | 0.6933 | 0.3924 | 0.3861 |
80
+ | 0.0091 | 10.71 | 300 | 0.7067 | 0.3991 | 0.3932 |
 
 
 
 
 
 
 
 
81
 
82
 
83
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cebf112b1eff3ebc1121e2ffcf99394f09d8a35f0e4c79b58718828ff520b2a5
3
  size 151097459
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ff0e1a73f22f8b7ed4abd67b8bcb35f26457dfa057c6959342a179b387ec355
3
  size 151097459
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cef9eff0f26cfd56f369720eef2f6ea99d2284cf658ec29b2035ac484081295c
3
  size 4143
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4444a045e5cb2c9dc13beb1626ef77a77ca978e3f8e15b3a89c61414ce5f069c
3
  size 4143