<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="abduljalil-etal-2026-arabic">
    <titleInfo>
        <title>Arabic-Adapted One-Step Speech-to-Diacritized ASR: Evaluation and Error Analysis</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Osamah</namePart>
        <namePart type="given">A</namePart>
        <namePart type="given">I</namePart>
        <namePart type="family">Abduljalil</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dalal</namePart>
        <namePart type="family">Ali</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Razan</namePart>
        <namePart type="given">A</namePart>
        <namePart type="family">Bajaman</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Abdullah</namePart>
        <namePart type="given">I</namePart>
        <namePart type="family">Alharbi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-03</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Rabat, Morocco</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Arabic diacritics encode phonetic information essential for pronunciation, disambiguation, and downstream applications, yet most Arabic ASR systems generate undiacritized output. In this work, we study direct speech-to-diacritized-text recognition using a single-stage ASR pipeline that predicts diacritics jointly with Arabic letters, without text-based post-processing. We evaluate two Arabic-adapted ASR architectures—wav2vec 2.0 XLSR-53 and Whisper-base—under a unified experimental setup on the ClArTTS Classical Arabic dataset. Performance is assessed using surface and lexical WER/CER alongside diacritic error rate (DER) to disentangle base transcription accuracy from diacritic realization. Our results show that Arabic-adapted wav2vec 2.0 achieves substantially lower diacritic error rates than Whisper, indicating stronger exploitation of acoustic cues relevant to vowelization. We further analyze the effect of decoding strategy and provide a detailed breakdown of diacritic errors, highlighting challenges associated with short vowels and morphosyntactic markers. These findings underscore the importance of model architecture and Arabic-specific adaptation for accurate diacritized Arabic ASR.</abstract>
    <identifier type="citekey">abduljalil-etal-2026-arabic</identifier>
    <location>
        <url>https://aclanthology.org/2026.abjadnlp-1.43/</url>
    </location>
    <part>
        <date>2026-03</date>
        <extent unit="page">
            <start>371</start>
            <end>379</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Arabic-Adapted One-Step Speech-to-Diacritized ASR: Evaluation and Error Analysis
%A Abduljalil, Osamah A. I.
%A Ali, Dalal
%A Bajaman, Razan A.
%A Alharbi, Abdullah I.
%S Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%F abduljalil-etal-2026-arabic
%X Arabic diacritics encode phonetic information essential for pronunciation, disambiguation, and downstream applications, yet most Arabic ASR systems generate undiacritized output. In this work, we study direct speech-to-diacritized-text recognition using a single-stage ASR pipeline that predicts diacritics jointly with Arabic letters, without text-based post-processing. We evaluate two Arabic-adapted ASR architectures—wav2vec 2.0 XLSR-53 and Whisper-base—under a unified experimental setup on the ClArTTS Classical Arabic dataset. Performance is assessed using surface and lexical WER/CER alongside diacritic error rate (DER) to disentangle base transcription accuracy from diacritic realization. Our results show that Arabic-adapted wav2vec 2.0 achieves substantially lower diacritic error rates than Whisper, indicating stronger exploitation of acoustic cues relevant to vowelization. We further analyze the effect of decoding strategy and provide a detailed breakdown of diacritic errors, highlighting challenges associated with short vowels and morphosyntactic markers. These findings underscore the importance of model architecture and Arabic-specific adaptation for accurate diacritized Arabic ASR.
%U https://aclanthology.org/2026.abjadnlp-1.43/
%P 371-379

Markdown (Informal)

[Arabic-Adapted One-Step Speech-to-Diacritized ASR: Evaluation and Error Analysis](https://aclanthology.org/2026.abjadnlp-1.43/) (Abduljalil et al., AbjadNLP 2026)

• Arabic-Adapted One-Step Speech-to-Diacritized ASR: Evaluation and Error Analysis (Abduljalil et al., AbjadNLP 2026)

ACL

• Osamah A. I. Abduljalil, Dalal Ali, Razan A. Bajaman, and Abdullah I. Alharbi. 2026. Arabic-Adapted One-Step Speech-to-Diacritized ASR: Evaluation and Error Analysis. In Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script, pages 371–379, Rabat, Morocco. Association for Computational Linguistics.