Hugging Face's logo

controngo
/
whisper-tiny-cv-de

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
German
whisper
hf-asr-leaderboard
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
 Metrics Community
controngo commited on
Commit
088b96a
·
verified ·
1 Parent(s): 14ba279

End of training

Browse files
Files changed (1) hide show
  1. README.md +14 -14
README.md CHANGED
@@ -25,7 +25,7 @@ model-index:
25
  metrics:
26
  - name: Wer
27
  type: wer
28
- value: 82.95482295482296
29
  ---
30
 
31
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -35,8 +35,8 @@ should probably proofread and complete it, then remove this comment. -->
35
 
36
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Common Voice 11.0 de 5% dataset.
37
  It achieves the following results on the evaluation set:
38
- - Loss: 0.7079
39
- - Wer: 82.9548
40
 
41
  ## Model description
42
 
@@ -55,28 +55,28 @@ More information needed
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
- - learning_rate: 1e-05
59
  - train_batch_size: 16
60
  - eval_batch_size: 8
61
  - seed: 42
62
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
  - lr_scheduler_type: linear
64
- - lr_scheduler_warmup_steps: 500
65
- - training_steps: 4000
66
  - mixed_precision_training: Native AMP
67
 
68
  ### Training results
69
 
70
  | Training Loss | Epoch | Step | Validation Loss | Wer |
71
  |:-------------:|:------:|:----:|:---------------:|:-------:|
72
- | 0.6087 | 0.4505 | 500 | 0.8082 | 78.1441 |
73
- | 0.527 | 0.9009 | 1000 | 0.7504 | 82.5641 |
74
- | 0.4246 | 1.3514 | 1500 | 0.7281 | 78.5104 |
75
- | 0.4013 | 1.8018 | 2000 | 0.7131 | 81.7338 |
76
- | 0.2809 | 2.2523 | 2500 | 0.7136 | 90.3053 |
77
- | 0.3299 | 2.7027 | 3000 | 0.7078 | 87.2527 |
78
- | 0.2573 | 3.1532 | 3500 | 0.7070 | 89.4261 |
79
- | 0.2346 | 3.6036 | 4000 | 0.7079 | 82.9548 |
80
 
81
 
82
  ### Framework versions
 
25
  metrics:
26
  - name: Wer
27
  type: wer
28
+ value: 72.91819291819291
29
  ---
30
 
31
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
35
 
36
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Common Voice 11.0 de 5% dataset.
37
  It achieves the following results on the evaluation set:
38
+ - Loss: 0.7117
39
+ - Wer: 72.9182
40
 
41
  ## Model description
42
 
 
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
+ - learning_rate: 1.35e-05
59
  - train_batch_size: 16
60
  - eval_batch_size: 8
61
  - seed: 42
62
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
  - lr_scheduler_type: linear
64
+ - lr_scheduler_warmup_steps: 250
65
+ - training_steps: 2000
66
  - mixed_precision_training: Native AMP
67
 
68
  ### Training results
69
 
70
  | Training Loss | Epoch | Step | Validation Loss | Wer |
71
  |:-------------:|:------:|:----:|:---------------:|:-------:|
72
+ | 0.6076 | 0.2252 | 250 | 0.8347 | 76.3126 |
73
+ | 0.5955 | 0.4505 | 500 | 0.7893 | 79.1697 |
74
+ | 0.5179 | 0.6757 | 750 | 0.7593 | 82.1978 |
75
+ | 0.5189 | 0.9009 | 1000 | 0.7370 | 73.0159 |
76
+ | 0.3644 | 1.1261 | 1250 | 0.7254 | 84.1270 |
77
+ | 0.394 | 1.3514 | 1500 | 0.7183 | 73.4066 |
78
+ | 0.3672 | 1.5766 | 1750 | 0.7152 | 73.1136 |
79
+ | 0.3751 | 1.8018 | 2000 | 0.7117 | 72.9182 |
80
 
81
 
82
  ### Framework versions