File size: 2,489 Bytes
5fd45b5
 
 
 
 
 
 
 
 
 
 
5af4f5a
5fd45b5
0424398
5fd45b5
 
0424398
5af4f5a
5fd45b5
 
 
 
 
 
 
5af4f5a
0424398
5af4f5a
5fd45b5
 
 
 
 
e3f91a9
5fd45b5
0424398
5fd45b5
0424398
 
5fd45b5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0424398
3bfd46c
 
5fd45b5
0424398
 
5fd45b5
3bfd46c
0424398
3bfd46c
5fd45b5
 
 
 
 
 
0424398
 
 
 
 
 
 
 
 
 
5fd45b5
 
 
 
 
 
 
 
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
85
86
87
88
89
90
91
92
---
language:
- tr
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
base_model: openai/whisper-small
model-index:
- name: Whisper Small Tr - Canberk Kandemir
  results:
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: Common Voice 11.0
      type: mozilla-foundation/common_voice_11_0
      config: tr
      split: None
      args: 'config: tr, split: test'
    metrics:
    - type: wer
      value: 43.06339873086104
      name: Wer
---

<!-- 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. -->

# Whisper Small Tr 

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5432
- Wer: 43.0634

## 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: 7e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- training_steps: 5000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.4399        | 0.44  | 500  | 0.6307          | 61.0351 |
| 0.4322        | 0.89  | 1000 | 0.6820          | 58.6909 |
| 0.2857        | 1.33  | 1500 | 0.6496          | 54.3867 |
| 0.2839        | 1.77  | 2000 | 0.6088          | 49.6497 |
| 0.1467        | 2.21  | 2500 | 0.5813          | 47.3346 |
| 0.1268        | 2.66  | 3000 | 0.5647          | 46.1315 |
| 0.0711        | 3.1   | 3500 | 0.5532          | 44.8196 |
| 0.0658        | 3.54  | 4000 | 0.5444          | 43.4670 |
| 0.0601        | 3.99  | 4500 | 0.5372          | 43.4146 |
| 0.0304        | 4.43  | 5000 | 0.5432          | 43.0634 |


### Framework versions

- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2