---
language:
- ara
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- AsemBadr/Al-Rahman
model-index:
- name: Whisper Small testing
  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 Small testing

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the testing dataset.

## 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: 1e-05
- train_batch_size: 16
- 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: 500
- training_steps: 500

### Training results



### Framework versions

- Transformers 4.38.0.dev0
- Pytorch 2.2.0+cpu
- Datasets 2.17.0
- Tokenizers 0.15.2