Hugging Face's logo

ezrab
/
whisper-tiny-minds14

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
whisper
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
 Metrics Community
ezrab commited on
Commit
716cc59
·
verified ·
1 Parent(s): 1925b96

End of training

Browse files
Files changed (3) hide show
  1. README.md +17 -18
  2. model.safetensors +1 -1
  3. training_args.bin +2 -2
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Wer
25
  type: wer
26
- value: 0.33116883116883117
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,9 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.7091
37
- - Wer Ortho: 0.3282
38
- - Wer: 0.3312
39
 
40
  ## Model description
41
 
@@ -57,32 +57,31 @@ The following hyperparameters were used during training:
57
  - learning_rate: 2e-05
58
  - train_batch_size: 16
59
  - eval_batch_size: 16
60
- - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: constant_with_warmup
63
  - lr_scheduler_warmup_steps: 50
64
  - training_steps: 500
65
- - mixed_precision_training: Native AMP
66
 
67
  ### Training results
68
 
69
  | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
70
  |:-------------:|:-------:|:----:|:---------------:|:---------:|:------:|
71
- | 1.1248 | 1.7857 | 50 | 0.5870 | 0.3868 | 0.3808 |
72
- | 0.2048 | 3.5714 | 100 | 0.5389 | 0.3152 | 0.3129 |
73
- | 0.0316 | 5.3571 | 150 | 0.6006 | 0.3233 | 0.3264 |
74
- | 0.0083 | 7.1429 | 200 | 0.6418 | 0.3350 | 0.3359 |
75
- | 0.0025 | 8.9286 | 250 | 0.6459 | 0.3276 | 0.3329 |
76
- | 0.0012 | 10.7143 | 300 | 0.6761 | 0.3270 | 0.3294 |
77
- | 0.0007 | 12.5 | 350 | 0.6790 | 0.3319 | 0.3335 |
78
- | 0.0004 | 14.2857 | 400 | 0.6967 | 0.3313 | 0.3329 |
79
- | 0.0008 | 16.0714 | 450 | 0.7005 | 0.3319 | 0.3347 |
80
- | 0.0006 | 17.8571 | 500 | 0.7091 | 0.3282 | 0.3312 |
81
 
82
 
83
  ### Framework versions
84
 
85
  - Transformers 4.44.2
86
- - Pytorch 2.4.1+cu121
87
- - Datasets 3.0.1
88
  - Tokenizers 0.19.1
 
23
  metrics:
24
  - name: Wer
25
  type: wer
26
+ value: 0.28512396694214875
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
33
 
34
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.7381
37
+ - Wer Ortho: 0.2850
38
+ - Wer: 0.2851
39
 
40
  ## Model description
41
 
 
57
  - learning_rate: 2e-05
58
  - train_batch_size: 16
59
  - eval_batch_size: 16
60
+ - seed: 0
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: constant_with_warmup
63
  - lr_scheduler_warmup_steps: 50
64
  - training_steps: 500
 
65
 
66
  ### Training results
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
69
  |:-------------:|:-------:|:----:|:---------------:|:---------:|:------:|
70
+ | 0.9105 | 1.7857 | 50 | 0.6418 | 0.4115 | 0.3937 |
71
+ | 0.2535 | 3.5714 | 100 | 0.5773 | 0.3337 | 0.3164 |
72
+ | 0.0887 | 5.3571 | 150 | 0.6295 | 0.3368 | 0.3182 |
73
+ | 0.0288 | 7.1429 | 200 | 0.6449 | 0.3381 | 0.3211 |
74
+ | 0.0198 | 8.9286 | 250 | 0.6932 | 0.4170 | 0.4203 |
75
+ | 0.0092 | 10.7143 | 300 | 0.6835 | 0.3152 | 0.3058 |
76
+ | 0.0134 | 12.5 | 350 | 0.7404 | 0.3288 | 0.3264 |
77
+ | 0.0096 | 14.2857 | 400 | 0.7067 | 0.3374 | 0.3312 |
78
+ | 0.0073 | 16.0714 | 450 | 0.7303 | 0.3122 | 0.3081 |
79
+ | 0.0056 | 17.8571 | 500 | 0.7381 | 0.2850 | 0.2851 |
80
 
81
 
82
  ### Framework versions
83
 
84
  - Transformers 4.44.2
85
+ - Pytorch 2.4.1
86
+ - Datasets 3.0.0
87
  - Tokenizers 0.19.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6238ef4eb067ba2bc53d0f894781e353d812cecbd714588d538841b087774384
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4144c808ba5313c92245706fdea3810a07eeda7e62c82bc82c5ae6cf4c76f7bb
3
  size 151061672
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7b0837bb40134daa8e9178f8e487d8f5ee696e958a0ea6e9956bed55483e7152
3
- size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c3bf494e72e5ba5a62cd7ef40eac2fc70b5f112684f0a9ef52a67ce18e95923
3
+ size 5304