---
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: 0.0168
- Wer: 0.0933
- Cer: 0.0351

## 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: 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    |
|:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
| 0.0099        | 1.0     | 157  | 0.0100          | 0.1046 | 0.0351 |
| 0.0082        | 2.0     | 314  | 0.0097          | 0.1061 | 0.0376 |
| 0.0067        | 3.0     | 471  | 0.0107          | 0.1294 | 0.0416 |
| 0.0046        | 4.0     | 628  | 0.0124          | 0.1352 | 0.0468 |
| 0.0034        | 5.0     | 785  | 0.0136          | 0.1281 | 0.0458 |
| 0.0027        | 6.0     | 942  | 0.0138          | 0.1348 | 0.0499 |
| 0.0019        | 7.0     | 1099 | 0.0137          | 0.1202 | 0.0408 |
| 0.0013        | 8.0     | 1256 | 0.0139          | 0.1169 | 0.0410 |
| 0.0009        | 9.0     | 1413 | 0.0144          | 0.1182 | 0.0430 |
| 0.0008        | 10.0    | 1570 | 0.0154          | 0.1117 | 0.0416 |
| 0.0006        | 11.0    | 1727 | 0.0156          | 0.1176 | 0.0428 |
| 0.0004        | 12.0    | 1884 | 0.0154          | 0.1026 | 0.0357 |
| 0.0002        | 13.0    | 2041 | 0.0154          | 0.1099 | 0.0409 |
| 0.0001        | 14.0    | 2198 | 0.0159          | 0.1050 | 0.0384 |
| 0.0001        | 15.0    | 2355 | 0.0153          | 0.1033 | 0.0414 |
| 0.0001        | 16.0    | 2512 | 0.0149          | 0.0923 | 0.0348 |
| 0.0           | 17.0    | 2669 | 0.0153          | 0.0914 | 0.0362 |
| 0.0           | 18.0    | 2826 | 0.0154          | 0.0892 | 0.0330 |
| 0.0           | 19.0    | 2983 | 0.0165          | 0.0940 | 0.0367 |
| 0.0           | 19.8768 | 3120 | 0.0161          | 0.0854 | 0.0322 |


### Framework versions

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