File size: 1,582 Bytes
1ea0309
 
 
 
 
 
 
6dc6f14
1ea0309
 
 
 
 
 
 
 
 
 
6dc6f14
1ea0309
6dc6f14
1ea0309
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6dc6f14
1ea0309
 
 
 
 
6dc6f14
 
 
 
 
1ea0309
 
 
 
 
 
 
 
 
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
---
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
datasets:
- common_voice_13_0
base_model: openai/whisper-tiny
model-index:
- name: arabic_speech_ASR_Whisper
  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. -->

# arabic_speech_ASR_Whisper

This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the common_voice_13_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4170

## 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: 0.001
- train_batch_size: 6
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.6558        | 1.0   | 1334 | 0.6054          |
| 0.4922        | 2.0   | 2668 | 0.5290          |
| 0.4324        | 3.0   | 4002 | 0.4767          |
| 0.3572        | 4.0   | 5336 | 0.4277          |
| 0.2179        | 5.0   | 6670 | 0.4170          |


### Framework versions

- PEFT 0.9.0
- Transformers 4.38.1
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2