بِسْمِ اللَّهِ الرَّحْمَٰنِ الرَّحِيمِ
In the name of Allah, the Most Gracious, the Most Merciful
Word-level timing data for 3 Quran reciters. Each JSON contains 6,236 ayahs with precise word timestamps.
OpenAI Whisper large-v3 with dynamic time warping alignment against Uthmani text.
Created using Quran_Aligner_Colab.ipynb (Google Colab) or kaggle.ipynb (Kaggle with T4 GPU). Audio from everyayah.com.
A specialized variant of the aligner for reciters who skip the Bismillah at the beginning of surahs in their audio recordings. Standard alignment expects Bismillah in the audio, causing misalignment when it's absent. This notebook adjusts the reference text to match the actual audio content, ensuring accurate word-level timestamps for these reciters.
[
{
"surah": 1,
"ayah": 1,
"segments": [[0, 1, 0, 780], [1, 2, 780, 1120], ...]
}
]Each segment: [word_start_idx, word_end_idx, start_ms, end_ms]
- Abdullah Basfar
- Ahmed ibn Ali al-Ajamy
- Ahmed Neana
- Akram AlAlaqimy
- Ali Jaber
- Ayman Sowaid
- Fares Abbad
- Ghamadi
- Hudhaify
- Ibrahim Akhdar
- Khaalid al-Qahtaanee
- Maher Al-Muaiqly
- Muhammad Ayyoub
- Muhammad Jibreel
- Muhsin Al-Qasim
- Nasser Alqatami
- Salaah Bukhatir
- Salah Al-Budair
- Warsh (Yassin al-Jazaery)
Generated by Yazan Baker using OpenAI Whisper.
- Quran text: Tanzil.net (Uthmani script)
- Audio: EveryAyah.com
- Speech recognition: OpenAI Whisper
Free for personal, educational, and commercial use. Please credit the original audio sources.
Generated: January 2026