@inproceedings{ashraf-etal-2026-bulbul,
title = "Bulbul: A Dataset for Dialectal {A}rabic Speech Recognition",
author = "Ashraf, Ahmed and
Alansari, Aisha and
Al Abbas, Fadel and
Almarwani, Nada and
Aloufi, Samah and
Ezzini, Saad and
Al-shaibani, Maged S. and
Dalaq, Doaa and
Elmadany, AbdelRahim A. and
Abdul-Mageed, Muhammad and
Trigui, Mohamed Mehdi and
Refai, Dania and
Refai, Layan and
Akrout, Mohamed and
Jarrar, Mustafa and
Al-Khatib, Wasfi G. and
Dalaq, Alaa and
El-Nakla, Darin and
Abdaljalil, Samir and
Al-Fakih, Abdulrahman and
Zeghib, Nour El Imane and
Redah, Moussa and
Chafik, Salmane and
El-Attar, Mohamed and
Grati, Rima and
Kohail, Sarah and
Alkhorasani, Malak and
Al Safwan, Khadijah and
Mudhaffar, Ismail Mohmmed and
Altam, Ali and
Al-Shaikh, Ahmed and
Saeed, Adnan Mahyoub Ahmed and
Luqman, Hamzah",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1032/",
pages = "23702--23720",
ISBN = "979-8-89176-459-0",
abstract = "Arabic automatic speech recognition (ASR) faces unique challenges due to diglossia, extensive regional dialect variation, and limited systematically balanced speech resources. Existing speech datasets often focus on single dialects or large-scale broadcast/web data, leading to trade-offs between linguistic diversity and annotation quality. We present BULBUL, a multi-dialect Arabic ASR dataset collected from 275 speakers across 11 Arab countries. BULBUL includes structured dialect and sub-dialect coverage, as well as recordings of classical Arabic and modern standard Arabic spoken by participants in their native dialectal accents to support accent-aware modeling. The quality of the recordings was ensured through a two-level human verification process. We further benchmark a range of recent ASR systems, establishing strong baselines for modern dialectal and accented Arabic ASR."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ashraf-etal-2026-bulbul">
<titleInfo>
<title>Bulbul: A Dataset for Dialectal Arabic Speech Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="family">Ashraf</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aisha</namePart>
<namePart type="family">Alansari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fadel</namePart>
<namePart type="family">Al Abbas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nada</namePart>
<namePart type="family">Almarwani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samah</namePart>
<namePart type="family">Aloufi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Saad</namePart>
<namePart type="family">Ezzini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maged</namePart>
<namePart type="given">S</namePart>
<namePart type="family">Al-shaibani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Doaa</namePart>
<namePart type="family">Dalaq</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">AbdelRahim</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Elmadany</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="family">Abdul-Mageed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="given">Mehdi</namePart>
<namePart type="family">Trigui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dania</namePart>
<namePart type="family">Refai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Layan</namePart>
<namePart type="family">Refai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="family">Akrout</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mustafa</namePart>
<namePart type="family">Jarrar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wasfi</namePart>
<namePart type="given">G</namePart>
<namePart type="family">Al-Khatib</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alaa</namePart>
<namePart type="family">Dalaq</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Darin</namePart>
<namePart type="family">El-Nakla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samir</namePart>
<namePart type="family">Abdaljalil</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdulrahman</namePart>
<namePart type="family">Al-Fakih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nour</namePart>
<namePart type="given">El</namePart>
<namePart type="given">Imane</namePart>
<namePart type="family">Zeghib</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Moussa</namePart>
<namePart type="family">Redah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salmane</namePart>
<namePart type="family">Chafik</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="family">El-Attar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rima</namePart>
<namePart type="family">Grati</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sarah</namePart>
<namePart type="family">Kohail</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Malak</namePart>
<namePart type="family">Alkhorasani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khadijah</namePart>
<namePart type="family">Al Safwan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ismail</namePart>
<namePart type="given">Mohmmed</namePart>
<namePart type="family">Mudhaffar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ali</namePart>
<namePart type="family">Altam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="family">Al-Shaikh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adnan</namePart>
<namePart type="given">Mahyoub</namePart>
<namePart type="given">Ahmed</namePart>
<namePart type="family">Saeed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hamzah</namePart>
<namePart type="family">Luqman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Arabic automatic speech recognition (ASR) faces unique challenges due to diglossia, extensive regional dialect variation, and limited systematically balanced speech resources. Existing speech datasets often focus on single dialects or large-scale broadcast/web data, leading to trade-offs between linguistic diversity and annotation quality. We present BULBUL, a multi-dialect Arabic ASR dataset collected from 275 speakers across 11 Arab countries. BULBUL includes structured dialect and sub-dialect coverage, as well as recordings of classical Arabic and modern standard Arabic spoken by participants in their native dialectal accents to support accent-aware modeling. The quality of the recordings was ensured through a two-level human verification process. We further benchmark a range of recent ASR systems, establishing strong baselines for modern dialectal and accented Arabic ASR.</abstract>
<identifier type="citekey">ashraf-etal-2026-bulbul</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1032/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>23702</start>
<end>23720</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Bulbul: A Dataset for Dialectal Arabic Speech Recognition
%A Ashraf, Ahmed
%A Alansari, Aisha
%A Al Abbas, Fadel
%A Almarwani, Nada
%A Aloufi, Samah
%A Ezzini, Saad
%A Al-shaibani, Maged S.
%A Dalaq, Doaa
%A Elmadany, AbdelRahim A.
%A Abdul-Mageed, Muhammad
%A Trigui, Mohamed Mehdi
%A Refai, Dania
%A Refai, Layan
%A Akrout, Mohamed
%A Jarrar, Mustafa
%A Al-Khatib, Wasfi G.
%A Dalaq, Alaa
%A El-Nakla, Darin
%A Abdaljalil, Samir
%A Al-Fakih, Abdulrahman
%A Zeghib, Nour El Imane
%A Redah, Moussa
%A Chafik, Salmane
%A El-Attar, Mohamed
%A Grati, Rima
%A Kohail, Sarah
%A Alkhorasani, Malak
%A Al Safwan, Khadijah
%A Mudhaffar, Ismail Mohmmed
%A Altam, Ali
%A Al-Shaikh, Ahmed
%A Saeed, Adnan Mahyoub Ahmed
%A Luqman, Hamzah
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F ashraf-etal-2026-bulbul
%X Arabic automatic speech recognition (ASR) faces unique challenges due to diglossia, extensive regional dialect variation, and limited systematically balanced speech resources. Existing speech datasets often focus on single dialects or large-scale broadcast/web data, leading to trade-offs between linguistic diversity and annotation quality. We present BULBUL, a multi-dialect Arabic ASR dataset collected from 275 speakers across 11 Arab countries. BULBUL includes structured dialect and sub-dialect coverage, as well as recordings of classical Arabic and modern standard Arabic spoken by participants in their native dialectal accents to support accent-aware modeling. The quality of the recordings was ensured through a two-level human verification process. We further benchmark a range of recent ASR systems, establishing strong baselines for modern dialectal and accented Arabic ASR.
%U https://aclanthology.org/2026.emnlp-main.1032/
%P 23702-23720
Markdown (Informal)
[Bulbul: A Dataset for Dialectal Arabic Speech Recognition](https://aclanthology.org/2026.emnlp-main.1032/) (Ashraf et al., EMNLP 2026)
ACL
- Ahmed Ashraf, Aisha Alansari, Fadel Al Abbas, Nada Almarwani, Samah Aloufi, Saad Ezzini, Maged S. Al-shaibani, Doaa Dalaq, AbdelRahim A. Elmadany, Muhammad Abdul-Mageed, Mohamed Mehdi Trigui, Dania Refai, Layan Refai, Mohamed Akrout, Mustafa Jarrar, Wasfi G. Al-Khatib, Alaa Dalaq, Darin El-Nakla, Samir Abdaljalil, Abdulrahman Al-Fakih, Nour El Imane Zeghib, Moussa Redah, Salmane Chafik, Mohamed El-Attar, Rima Grati, Sarah Kohail, Malak Alkhorasani, Khadijah Al Safwan, Ismail Mohmmed Mudhaffar, Ali Altam, Ahmed Al-Shaikh, Adnan Mahyoub Ahmed Saeed, and Hamzah Luqman. 2026. Bulbul: A Dataset for Dialectal Arabic Speech Recognition. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 23702–23720, Budapest, Hungary. Association for Computational Linguistics.