---
library_name: transformers
language:
- ar
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper tiny AR - BH
  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. -->

# Whisper tiny AR - BH

This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the quran-ayat-speech-to-text dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0058
- Wer: 0.0698
- Cer: 0.0282

## 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: 5e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 20
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Wer    | Cer    |
|:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
| 0.0068        | 1.0     | 157  | 0.0060          | 0.0635 | 0.0258 |
| 0.005         | 2.0     | 314  | 0.0058          | 0.0713 | 0.0324 |
| 0.0055        | 3.0     | 471  | 0.0057          | 0.0624 | 0.0255 |
| 0.0064        | 4.0     | 628  | 0.0057          | 0.0628 | 0.0264 |
| 0.0042        | 5.0     | 785  | 0.0059          | 0.0644 | 0.0272 |
| 0.0031        | 6.0     | 942  | 0.0061          | 0.0646 | 0.0258 |
| 0.0022        | 7.0     | 1099 | 0.0063          | 0.0664 | 0.0271 |
| 0.0024        | 8.0     | 1256 | 0.0067          | 0.0652 | 0.0261 |
| 0.0017        | 9.0     | 1413 | 0.0071          | 0.0652 | 0.0257 |
| 0.0011        | 10.0    | 1570 | 0.0075          | 0.0662 | 0.0268 |
| 0.0006        | 11.0    | 1727 | 0.0078          | 0.0664 | 0.0267 |
| 0.0006        | 12.0    | 1884 | 0.0082          | 0.1468 | 0.0459 |
| 0.0006        | 13.0    | 2041 | 0.0084          | 0.1486 | 0.0543 |
| 0.0003        | 14.0    | 2198 | 0.0086          | 0.1486 | 0.0549 |
| 0.0003        | 15.0    | 2355 | 0.0087          | 0.1497 | 0.0546 |
| 0.0001        | 16.0    | 2512 | 0.0089          | 0.1479 | 0.0538 |
| 0.0003        | 17.0    | 2669 | 0.0089          | 0.1495 | 0.0542 |
| 0.0001        | 18.0    | 2826 | 0.0090          | 0.2304 | 0.0930 |
| 0.0001        | 19.0    | 2983 | 0.0094          | 0.1203 | 0.0502 |
| 0.0001        | 19.8768 | 3120 | 0.0090          | 0.2286 | 0.0921 |


### Framework versions

- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0