@inproceedings{stamou-etal-2026-first,
title = "First Steps in {ASR} for {C}ypriot {G}reek: Challenges and Insights",
author = "Stamou, Vivian and
Armostis, Spyros and
Klimi, Antigoni and
Paraskevopoulos, Georgios and
Katsouros, Vassilis and
Anastasopoulos, Antonios",
editor = "Anastasopoulos, Antonis and
Markantonatou, Stella and
Ralli, Angela and
Zampieri, Marcos and
Bompolas, Stavros and
Stamou, Vivian",
booktitle = "Proceedings of the First Workshop on Dialects in {NLP} {---} A Resource Perspective",
month = may,
year = "2026",
address = "Palma de Mallorca",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.dialres-1.30/",
doi = "10.63317/2iznmge7hdtx",
pages = "308--314",
abstract = "This paper presents the first automatic speech recognition (ASR) system for Cypriot Greek, a non-standardized variety of Modern Greek with distinctive phonological, lexical, and orthographic characteristics. We adapt Whisper, a state-of-the-art multilingual ASR model, to Cypriot Greek through fine-tuning on the mozilla common voice spontaneous speech dataset for Cypriot Greek. The phonological and lexical divergence of Cypriot Greek from Standard Modern Greek poses significant challenges for mainstream ASR, particularly under conditions of limited training data and dialectal variation. Results demonstrate that whisper-medium achieved a best word error rate (WER) of 37.85{\%}, while whisper-large-v3 consistently outperformed it, reaching a minimum WER of 33.93{\%}. In the light of these findings, increased model size, combined with targeted fine-tuning on normalized dialectical data, significantly improves recognition accuracy, indicating that careful handling of orthographic and dialectical variation provides an effective path for ASR adaptation to low-resource varieties."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="stamou-etal-2026-first">
<titleInfo>
<title>First Steps in ASR for Cypriot Greek: Challenges and Insights</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vivian</namePart>
<namePart type="family">Stamou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Spyros</namePart>
<namePart type="family">Armostis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antigoni</namePart>
<namePart type="family">Klimi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Georgios</namePart>
<namePart type="family">Paraskevopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vassilis</namePart>
<namePart type="family">Katsouros</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonios</namePart>
<namePart type="family">Anastasopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the First Workshop on Dialects in NLP — A Resource Perspective</title>
</titleInfo>
<name type="personal">
<namePart type="given">Antonis</namePart>
<namePart type="family">Anastasopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stella</namePart>
<namePart type="family">Markantonatou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Angela</namePart>
<namePart type="family">Ralli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcos</namePart>
<namePart type="family">Zampieri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stavros</namePart>
<namePart type="family">Bompolas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivian</namePart>
<namePart type="family">Stamou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Palma de Mallorca</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper presents the first automatic speech recognition (ASR) system for Cypriot Greek, a non-standardized variety of Modern Greek with distinctive phonological, lexical, and orthographic characteristics. We adapt Whisper, a state-of-the-art multilingual ASR model, to Cypriot Greek through fine-tuning on the mozilla common voice spontaneous speech dataset for Cypriot Greek. The phonological and lexical divergence of Cypriot Greek from Standard Modern Greek poses significant challenges for mainstream ASR, particularly under conditions of limited training data and dialectal variation. Results demonstrate that whisper-medium achieved a best word error rate (WER) of 37.85%, while whisper-large-v3 consistently outperformed it, reaching a minimum WER of 33.93%. In the light of these findings, increased model size, combined with targeted fine-tuning on normalized dialectical data, significantly improves recognition accuracy, indicating that careful handling of orthographic and dialectical variation provides an effective path for ASR adaptation to low-resource varieties.</abstract>
<identifier type="citekey">stamou-etal-2026-first</identifier>
<identifier type="doi">10.63317/2iznmge7hdtx</identifier>
<location>
<url>https://aclanthology.org/2026.dialres-1.30/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>308</start>
<end>314</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T First Steps in ASR for Cypriot Greek: Challenges and Insights
%A Stamou, Vivian
%A Armostis, Spyros
%A Klimi, Antigoni
%A Paraskevopoulos, Georgios
%A Katsouros, Vassilis
%A Anastasopoulos, Antonios
%Y Anastasopoulos, Antonis
%Y Markantonatou, Stella
%Y Ralli, Angela
%Y Zampieri, Marcos
%Y Bompolas, Stavros
%Y Stamou, Vivian
%S Proceedings of the First Workshop on Dialects in NLP — A Resource Perspective
%D 2026
%8 May
%I Association for Computational Linguistics
%C Palma de Mallorca
%F stamou-etal-2026-first
%X This paper presents the first automatic speech recognition (ASR) system for Cypriot Greek, a non-standardized variety of Modern Greek with distinctive phonological, lexical, and orthographic characteristics. We adapt Whisper, a state-of-the-art multilingual ASR model, to Cypriot Greek through fine-tuning on the mozilla common voice spontaneous speech dataset for Cypriot Greek. The phonological and lexical divergence of Cypriot Greek from Standard Modern Greek poses significant challenges for mainstream ASR, particularly under conditions of limited training data and dialectal variation. Results demonstrate that whisper-medium achieved a best word error rate (WER) of 37.85%, while whisper-large-v3 consistently outperformed it, reaching a minimum WER of 33.93%. In the light of these findings, increased model size, combined with targeted fine-tuning on normalized dialectical data, significantly improves recognition accuracy, indicating that careful handling of orthographic and dialectical variation provides an effective path for ASR adaptation to low-resource varieties.
%R 10.63317/2iznmge7hdtx
%U https://aclanthology.org/2026.dialres-1.30/
%U https://doi.org/10.63317/2iznmge7hdtx
%P 308-314
Markdown (Informal)
[First Steps in ASR for Cypriot Greek: Challenges and Insights](https://aclanthology.org/2026.dialres-1.30/) (Stamou et al., DialRes 2026)
ACL
- Vivian Stamou, Spyros Armostis, Antigoni Klimi, Georgios Paraskevopoulos, Vassilis Katsouros, and Antonios Anastasopoulos. 2026. First Steps in ASR for Cypriot Greek: Challenges and Insights. In Proceedings of the First Workshop on Dialects in NLP — A Resource Perspective, pages 308–314, Palma de Mallorca. Association for Computational Linguistics.