@inproceedings{alhumid-etal-2026-abjad,
title = "Abjad {AI} at {NADI} 2026 shared task: Overcoming Data Scarcity in {A}rabic Speech Processing through Augmentation",
author = "Alhumid, Hatim and
Alharthi, Naif and
Alsaedi, Elaf S. and
Alasmary, Faris and
Ghannam, Ahmad and
Salamah, Anas Adel and
Sadah, Shouq and
Ghouti, Lahouari",
editor = "Abdul-Mageed, Muhammad and
Elsayed, Tamer and
Bouamor, Houda and
El-Beltagy, Samhaa and
Al-Khalifa, Hend and
Abdelali, Ahmed and
Elmadany, AbdelRahim and
El Mekki, Abdellah and
Darwish, Kareem and
Alkhereyf, Sakhar and
Luqman, Hamzah and
Khalifa, Salam and
Magdy, Samar and
Alyafeai, Zaid and
Hamed, Injy and
Zaghouani, Wajdi and
Zitouni, Imed",
booktitle = "Proceedings of The Fourth {A}rabic Natural Language Processing Conference: Shared Tasks",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.arabicnlp-sharedtasks.4/",
pages = "47--52",
ISBN = "979-8-89176-497-2",
abstract = "We describe our submission to three NADI 2026 subtasks: Tunisian Arabic{--}French{--}English code-switched ASR (Task 1.3), dialectal Arabic speech translation, and 20-way Arabic dialect identification (ADI). For code-switched ASR, we finetune a 2B-parameter FastConformer encoder-decoder with synthetic Tunisian speech, cross-lingual data, waveform augmentation, and checkpoint averaging, achieving 14.12{\%} WER and ranking first overall. For speech translation, finetuning Whisper Large-v3 yields strong in-domain results but poor generalization (BLEU 5.8 on blind test), illustrating overfitting under limited data. For ADI, a classifier on frozen pretrained ASR representations outperforms full/partial finetuning, reaching 42{\%} accuracy across 20 dialects. We highlight synthetic augmentation, checkpoint averaging, and encoder freezing as effective low-resource strategies."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="alhumid-etal-2026-abjad">
<titleInfo>
<title>Abjad AI at NADI 2026 shared task: Overcoming Data Scarcity in Arabic Speech Processing through Augmentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hatim</namePart>
<namePart type="family">Alhumid</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naif</namePart>
<namePart type="family">Alharthi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elaf</namePart>
<namePart type="given">S</namePart>
<namePart type="family">Alsaedi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Faris</namePart>
<namePart type="family">Alasmary</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmad</namePart>
<namePart type="family">Ghannam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anas</namePart>
<namePart type="given">Adel</namePart>
<namePart type="family">Salamah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shouq</namePart>
<namePart type="family">Sadah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lahouari</namePart>
<namePart type="family">Ghouti</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 Fourth Arabic Natural Language Processing Conference: Shared Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="family">Abdul-Mageed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tamer</namePart>
<namePart type="family">Elsayed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samhaa</namePart>
<namePart type="family">El-Beltagy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hend</namePart>
<namePart type="family">Al-Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="family">Abdelali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">AbdelRahim</namePart>
<namePart type="family">Elmadany</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdellah</namePart>
<namePart type="family">El Mekki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kareem</namePart>
<namePart type="family">Darwish</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakhar</namePart>
<namePart type="family">Alkhereyf</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hamzah</namePart>
<namePart type="family">Luqman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salam</namePart>
<namePart type="family">Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samar</namePart>
<namePart type="family">Magdy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zaid</namePart>
<namePart type="family">Alyafeai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Injy</namePart>
<namePart type="family">Hamed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wajdi</namePart>
<namePart type="family">Zaghouani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Imed</namePart>
<namePart type="family">Zitouni</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">979-8-89176-497-2</identifier>
</relatedItem>
<abstract>We describe our submission to three NADI 2026 subtasks: Tunisian Arabic–French–English code-switched ASR (Task 1.3), dialectal Arabic speech translation, and 20-way Arabic dialect identification (ADI). For code-switched ASR, we finetune a 2B-parameter FastConformer encoder-decoder with synthetic Tunisian speech, cross-lingual data, waveform augmentation, and checkpoint averaging, achieving 14.12% WER and ranking first overall. For speech translation, finetuning Whisper Large-v3 yields strong in-domain results but poor generalization (BLEU 5.8 on blind test), illustrating overfitting under limited data. For ADI, a classifier on frozen pretrained ASR representations outperforms full/partial finetuning, reaching 42% accuracy across 20 dialects. We highlight synthetic augmentation, checkpoint averaging, and encoder freezing as effective low-resource strategies.</abstract>
<identifier type="citekey">alhumid-etal-2026-abjad</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.4/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>47</start>
<end>52</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Abjad AI at NADI 2026 shared task: Overcoming Data Scarcity in Arabic Speech Processing through Augmentation
%A Alhumid, Hatim
%A Alharthi, Naif
%A Alsaedi, Elaf S.
%A Alasmary, Faris
%A Ghannam, Ahmad
%A Salamah, Anas Adel
%A Sadah, Shouq
%A Ghouti, Lahouari
%Y Abdul-Mageed, Muhammad
%Y Elsayed, Tamer
%Y Bouamor, Houda
%Y El-Beltagy, Samhaa
%Y Al-Khalifa, Hend
%Y Abdelali, Ahmed
%Y Elmadany, AbdelRahim
%Y El Mekki, Abdellah
%Y Darwish, Kareem
%Y Alkhereyf, Sakhar
%Y Luqman, Hamzah
%Y Khalifa, Salam
%Y Magdy, Samar
%Y Alyafeai, Zaid
%Y Hamed, Injy
%Y Zaghouani, Wajdi
%Y Zitouni, Imed
%S Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-497-2
%F alhumid-etal-2026-abjad
%X We describe our submission to three NADI 2026 subtasks: Tunisian Arabic–French–English code-switched ASR (Task 1.3), dialectal Arabic speech translation, and 20-way Arabic dialect identification (ADI). For code-switched ASR, we finetune a 2B-parameter FastConformer encoder-decoder with synthetic Tunisian speech, cross-lingual data, waveform augmentation, and checkpoint averaging, achieving 14.12% WER and ranking first overall. For speech translation, finetuning Whisper Large-v3 yields strong in-domain results but poor generalization (BLEU 5.8 on blind test), illustrating overfitting under limited data. For ADI, a classifier on frozen pretrained ASR representations outperforms full/partial finetuning, reaching 42% accuracy across 20 dialects. We highlight synthetic augmentation, checkpoint averaging, and encoder freezing as effective low-resource strategies.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.4/
%P 47-52
Markdown (Informal)
[Abjad AI at NADI 2026 shared task: Overcoming Data Scarcity in Arabic Speech Processing through Augmentation](https://aclanthology.org/2026.arabicnlp-sharedtasks.4/) (Alhumid et al., ArabicNLP 2026)
ACL
- Hatim Alhumid, Naif Alharthi, Elaf S. Alsaedi, Faris Alasmary, Ahmad Ghannam, Anas Adel Salamah, Shouq Sadah, and Lahouari Ghouti. 2026. Abjad AI at NADI 2026 shared task: Overcoming Data Scarcity in Arabic Speech Processing through Augmentation. In Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks, pages 47–52, Budapest, Hungary. Association for Computational Linguistics.