@inproceedings{thillainathan-etal-2026-talk,
title = "{T}a{LK}-Corpus: A Regionally Diverse Evaluation Set for {S}ri {L}ankan {T}amil Speech",
author = "Thillainathan, Adsajan and
Kanthakumar, Nishanthini and
Rasan, Nivethiga and
Sarveswaran, Kengatharaiyer",
editor = "Hosseini-Kivanani, Nina and
Brutti, Alessio and
Matassoni, Marco and
Dowerah, Sandipana and
Liga, Davide and
Schommer, Christoph",
booktitle = "Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis ({SPEAKABLE}) @ {LREC} 2026",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.speakable-1.21/",
doi = "10.63317/5edtmzph3wu4",
pages = "194--200",
abstract = "This paper introduces the TaLK Corpus, the first speech benchmark corpus for Sri Lankan Tamil Automatic Speech Recognition (ASR) covering speech from 22 administrative districts of Sri Lanka. The corpus contains 1 hour and 33 minutes of speech from 22 native speakers (one per district) and includes rich metadata on demographics, location history, recording conditions, and domain information, along with transcriptions in Tamil script and the International Phonetic Alphabet (IPA). Standardised preprocessing (16 kHz mono WAV format) and segmentation using Silero Voice Activity Detection (VAD) resulted in 1,214 utterances. All recordings were manually transcribed by trained linguists, and MD5-based file naming used to ensure data integrity and consistency. TaLK corpus enables district-wise benchmarking of ASR systems and supports dialect-sensitive evaluation. We establish baseline results for multilingual models (Whisper Large-V3 and Facebook{'}s MMS) in zero-shot settings. The evaluation reveals substantial performance disparities across districts, highlighting the impact of regional phonological variation in low-resource Sri Lankan Tamil. Although Whisper Large-V3 outperforms MMS overall, it shows considerable variability, with mean Word Error Rates ranging from 0.672 to 0.903 across districts. These findings demonstrate strong regional effects even within a single model. By releasing TaLK-Corpus under the CC-BY-NC 4.0 licence, we aim to support dialect-robust ASR research and foster inclusive speech technologies for Sri Lankan Tamil-speaking communities."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="thillainathan-etal-2026-talk">
<titleInfo>
<title>TaLK-Corpus: A Regionally Diverse Evaluation Set for Sri Lankan Tamil Speech</title>
</titleInfo>
<name type="personal">
<namePart type="given">Adsajan</namePart>
<namePart type="family">Thillainathan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nishanthini</namePart>
<namePart type="family">Kanthakumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nivethiga</namePart>
<namePart type="family">Rasan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kengatharaiyer</namePart>
<namePart type="family">Sarveswaran</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 Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nina</namePart>
<namePart type="family">Hosseini-Kivanani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessio</namePart>
<namePart type="family">Brutti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Matassoni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sandipana</namePart>
<namePart type="family">Dowerah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Davide</namePart>
<namePart type="family">Liga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christoph</namePart>
<namePart type="family">Schommer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper introduces the TaLK Corpus, the first speech benchmark corpus for Sri Lankan Tamil Automatic Speech Recognition (ASR) covering speech from 22 administrative districts of Sri Lanka. The corpus contains 1 hour and 33 minutes of speech from 22 native speakers (one per district) and includes rich metadata on demographics, location history, recording conditions, and domain information, along with transcriptions in Tamil script and the International Phonetic Alphabet (IPA). Standardised preprocessing (16 kHz mono WAV format) and segmentation using Silero Voice Activity Detection (VAD) resulted in 1,214 utterances. All recordings were manually transcribed by trained linguists, and MD5-based file naming used to ensure data integrity and consistency. TaLK corpus enables district-wise benchmarking of ASR systems and supports dialect-sensitive evaluation. We establish baseline results for multilingual models (Whisper Large-V3 and Facebook’s MMS) in zero-shot settings. The evaluation reveals substantial performance disparities across districts, highlighting the impact of regional phonological variation in low-resource Sri Lankan Tamil. Although Whisper Large-V3 outperforms MMS overall, it shows considerable variability, with mean Word Error Rates ranging from 0.672 to 0.903 across districts. These findings demonstrate strong regional effects even within a single model. By releasing TaLK-Corpus under the CC-BY-NC 4.0 licence, we aim to support dialect-robust ASR research and foster inclusive speech technologies for Sri Lankan Tamil-speaking communities.</abstract>
<identifier type="citekey">thillainathan-etal-2026-talk</identifier>
<identifier type="doi">10.63317/5edtmzph3wu4</identifier>
<location>
<url>https://aclanthology.org/2026.speakable-1.21/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>194</start>
<end>200</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TaLK-Corpus: A Regionally Diverse Evaluation Set for Sri Lankan Tamil Speech
%A Thillainathan, Adsajan
%A Kanthakumar, Nishanthini
%A Rasan, Nivethiga
%A Sarveswaran, Kengatharaiyer
%Y Hosseini-Kivanani, Nina
%Y Brutti, Alessio
%Y Matassoni, Marco
%Y Dowerah, Sandipana
%Y Liga, Davide
%Y Schommer, Christoph
%S Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F thillainathan-etal-2026-talk
%X This paper introduces the TaLK Corpus, the first speech benchmark corpus for Sri Lankan Tamil Automatic Speech Recognition (ASR) covering speech from 22 administrative districts of Sri Lanka. The corpus contains 1 hour and 33 minutes of speech from 22 native speakers (one per district) and includes rich metadata on demographics, location history, recording conditions, and domain information, along with transcriptions in Tamil script and the International Phonetic Alphabet (IPA). Standardised preprocessing (16 kHz mono WAV format) and segmentation using Silero Voice Activity Detection (VAD) resulted in 1,214 utterances. All recordings were manually transcribed by trained linguists, and MD5-based file naming used to ensure data integrity and consistency. TaLK corpus enables district-wise benchmarking of ASR systems and supports dialect-sensitive evaluation. We establish baseline results for multilingual models (Whisper Large-V3 and Facebook’s MMS) in zero-shot settings. The evaluation reveals substantial performance disparities across districts, highlighting the impact of regional phonological variation in low-resource Sri Lankan Tamil. Although Whisper Large-V3 outperforms MMS overall, it shows considerable variability, with mean Word Error Rates ranging from 0.672 to 0.903 across districts. These findings demonstrate strong regional effects even within a single model. By releasing TaLK-Corpus under the CC-BY-NC 4.0 licence, we aim to support dialect-robust ASR research and foster inclusive speech technologies for Sri Lankan Tamil-speaking communities.
%R 10.63317/5edtmzph3wu4
%U https://aclanthology.org/2026.speakable-1.21/
%U https://doi.org/10.63317/5edtmzph3wu4
%P 194-200
Markdown (Informal)
[TaLK-Corpus: A Regionally Diverse Evaluation Set for Sri Lankan Tamil Speech](https://aclanthology.org/2026.speakable-1.21/) (Thillainathan et al., SPEAKABLE 2026)
ACL
- Adsajan Thillainathan, Nishanthini Kanthakumar, Nivethiga Rasan, and Kengatharaiyer Sarveswaran. 2026. TaLK-Corpus: A Regionally Diverse Evaluation Set for Sri Lankan Tamil Speech. In Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026, pages 194–200, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).