Preprint

CARDAMOM: A Micro-Dialectal Arabic Speech Dataset for ASR

Bashar TalafhaThe University of British Columbia ,Samar M. MagdyThe University of British Columbia ,Aisha AlansariKing Fahd University of Petroleum and Minerals ,Alaa AlkhawaldehAl al-Bayt UniversityAbdurrahman JumaBirzeit University ,Sharaf MakahlehJordan University of Science and Technology ,Nour GamalBadr University in Cairo ,Omar AttiaBadr University in Cairo ,Hanaa KurdiTaibah University ,Najwa RizkBadr University in Cairo ,Maysa AnayaMAU-SouthernHessah AltimyatNorthern Border University ,Layal AlhazmiTaibah University ,Shumukh AlotaibiTaibah University ,Hajar AlhadarisJordan University of Science and Technology ,Rayan AlomariJordan University of Science and Technology ,Rahaf AlmalaqTaibah University ,Malak AlkhorasaniImam Abdulrahman Bin Faisal University ,Sara alghamdiTaibah University ,Rahaf AlshamraniImam Abdulrahman Bin Faisal University ,Nsrin AshrafEl Sewedy University of TechnologyIbrahim JaradatJordan University of Science and Technology ,Nada QardahjiJordan University of Science and Technology ,Yasmin ZaraketImperial College London ,Elmoukhtar BrahimInstitut Supérieur du Génie AppliquéSidi EbeidyInstitut Supérieur du Génie AppliquéOumoulmouminin MahmoudInstitut Supérieur du Numérique ,Yahjeb Bouha KhatratyInstitut Supérieur du Génie AppliquéMeya HarouneInstitut Supérieur du Numérique ,Mohammad GhaddarLebanese University ,Mohamad EldiranyLebanese University ,Rashed AlamoushJordan University of Science and Technology ,Tala ChhaytleAmerican University of Beirut ,Nuha AlbadiKing Fahd University of Petroleum and Minerals ,Yahya El HadjInstitut Supérieur du Numérique ,Hamzah LuqmanKing Fahd University of Petroleum and Minerals ,Fadi A. ZaraketMustafa JarrarHamad Bin Khalifa University ,Muhammad Abdul-MageedThe University of British Columbia ,

2026

Ochiq kirishYashil

ABI

Annotatsiya

We present Cardamom, a micro-dialectal Arabic speech dataset designed to support fine-grained evaluation and adaptation of automatic speech recognition (ASR) systems. Community-curated by native speakers familiar with the represented varieties, Cardamom contains approximately 40 hours of transcribed YouTube speech spanning 21 micro-dialects across Egypt, Jordan, Lebanon, Mauritania, Palestine, and Saudi Arabia. Each segment is annotated with one or more operational micro-dialect labels, code-switching information, and utterance-level perceived gender, enabling analysis of sub-country variation that is obscured by conventional country-level labels. We describe the collection and annotation process, motivate the micro-dialect inventory linguistically, and benchmark four multilingual ASR systems in zero-shot and adapted settings. The strongest zero-shot system obtains 43.47% aggregate WER, with particularly high error rates on Mauritanian and Lebanese varieties; adaptation on Cardamom reduces its WER to 35.21%. Audio-based identification experiments further show that the annotations provide a learnable prediction target, with a dedicated classifier reaching 85.57% accuracy on 21-way micro-dialect identification. Cardamom provides a resource for studying localized dialectal variation and developing Arabic speech systems with broader regional coverage.

Hali tarjima qilinmagan

Identifikatorlar

DOI: 10.48550/arxiv.2609.34481

Iqtiboslar va manbalar

0 ta iqtibos0 ta foydalanilgan manba