iarhamanwaar/quran-reciter-audio

---
language: ar
license: cc-by-4.0
task_categories:
- audio-classification
tags:
- quran
- arabic
- speaker-identification
- reciter-identification
- audio
size_categories:
- 1K<n<10K
---

# Quran Reciter Audio Dataset

Audio dataset of **362 Quran reciters** with ~25 MP3 files per reciter, totaling **8,792 files (~78 GB)**. Designed for training reciter/speaker identification models.

## Dataset Description

- **Reciters**: 362 unique Quran reciters
- **Total files**: 8,792 MP3 audio files
- **Average files per reciter**: ~24
- **Audio format**: MP3, various bitrates (128-320 kbps)
- **Content**: Quran surah recitations (each file is typically one surah or portion)
- **Source**: MP3Quran.net

## Dataset Structure

```
data/
Aadel_Alkalbani_(MP3_Quran)/
001.mp3
002.mp3
...
Aamer_Alkadhimi/
086.mp3
091.mp3
...
... (362 reciter directories)
```

Each directory is named after the reciter. Files are numbered by surah.

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("iarhamanwaar/quran-reciter-audio")
```

Or manually:
```python
from huggingface_hub import snapshot_download

snapshot_download("iarhamanwaar/quran-reciter-audio", repo_type="dataset", local_dir="./quran-data")
```

## Associated Model

A fine-tuned ECAPA-TDNN reciter identification model trained on this dataset is available at:
[iarhamanwaar/quran-reciter-id-ecapa](https://huggingface.co/iarhamanwaar/quran-reciter-id-ecapa)

## Statistics

| Metric | Value |
|--------|-------|
| Total reciters | 362 |
| Total audio files | 8,792 |
| Total size | ~78 GB |
| Min files per reciter | 2 |
| Max files per reciter | 33 |
| Audio format | MP3 |

## License & Attribution

This dataset is released under **CC-BY-4.0**. Audio recordings are sourced from [MP3Quran.net](https://www.mp3quran.net), a platform dedicated to the free distribution of Quran recitations.

Please attribute as:
```
Quran Reciter Audio Dataset by Arham Anwaar, sourced from MP3Quran.net
```

## Citation

```
@misc{quran-reciter-audio-2026,
title={Quran Reciter Audio Dataset: 362 Reciters for Speaker Identification},
author={Arham Anwaar},
year={2026},
url={https://huggingface.co/datasets/iarhamanwaar/quran-reciter-audio}
}
```