File size: 2,814 Bytes
8b842bf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
50bf7d0
 
 
8b842bf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
50bf7d0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8b842bf
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
---
language:
- de
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: openai/whisper-small
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# openai/whisper-small

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Hanhpt23/GermanMed-full dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6821
- Wer: 26.4630
- Cer: 15.4774

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 20

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     | Cer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
| 0.5026        | 1.0   | 194  | 0.5290          | 29.2811 | 19.4792 |
| 0.2354        | 2.0   | 388  | 0.5282          | 34.3515 | 22.4451 |
| 0.1144        | 3.0   | 582  | 0.5396          | 32.5825 | 21.4992 |
| 0.073         | 4.0   | 776  | 0.5676          | 32.2226 | 22.5456 |
| 0.0465        | 5.0   | 970  | 0.6049          | 26.3499 | 16.4094 |
| 0.0375        | 6.0   | 1164 | 0.6197          | 33.1791 | 21.1216 |
| 0.0213        | 7.0   | 1358 | 0.6250          | 30.1759 | 20.1462 |
| 0.0229        | 8.0   | 1552 | 0.6453          | 31.4718 | 19.4914 |
| 0.0118        | 9.0   | 1746 | 0.6510          | 23.1924 | 14.5627 |
| 0.0138        | 10.0  | 1940 | 0.6604          | 27.9235 | 17.4974 |
| 0.0081        | 11.0  | 2134 | 0.6546          | 26.3705 | 16.3176 |
| 0.0029        | 12.0  | 2328 | 0.6527          | 25.3625 | 15.1725 |
| 0.0028        | 13.0  | 2522 | 0.6712          | 22.6473 | 14.5384 |
| 0.0003        | 14.0  | 2716 | 0.6743          | 30.7004 | 18.0015 |
| 0.0002        | 15.0  | 2910 | 0.6752          | 27.2035 | 16.0248 |
| 0.0001        | 16.0  | 3104 | 0.6787          | 24.8277 | 15.1292 |
| 0.0001        | 17.0  | 3298 | 0.6803          | 26.6893 | 15.6541 |
| 0.0001        | 18.0  | 3492 | 0.6813          | 26.5864 | 15.6021 |
| 0.0001        | 19.0  | 3686 | 0.6819          | 26.4836 | 15.4964 |
| 0.0001        | 20.0  | 3880 | 0.6821          | 26.4630 | 15.4774 |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1