@inproceedings{s-etal-2026-wise,
title = "Wise@{D}ravidian{L}ang{T}ech 2026: Dialect-Aware {T}amil Speech Classification and Recognition via Cross-Pipeline Embedding Transfer",
author = "S, Ganesh Sundhar and
N, Hari Krishnan and
G, Gnanasabesan and
KP, Suriya and
G, Jyothish Lal",
editor = "Chakravarthi, Bharathi Raja and
Priyadharshini, Ruba and
Madasamy, Anand Kumar and
Thavareesan, Sajeetha and
Rajiakodi, Saranya and
Navaneethakrishnan, Subalalitha and
Chinnappa, Dhivya and
Palani, Balasubramanian and
Subramanian, Malliga and
Shanmugavadivel, Kogilavani and
Rajalakshmi, Ratnavel",
booktitle = "Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for {D}ravidian Languages",
month = jul,
year = "2026",
address = "Underline (Virtual)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.dravidianlangtech-1.71/",
pages = "447--452",
ISBN = "979-8-89176-401-9",
abstract = "This paper presents the **Wise** system for the shared task on dialect-based speech processing in Tamil, addressing two subtasks: **(1) four-way dialect region classification** (Northern, Southern, Western, Central), and **(2) dialectal Tamil ASR**. All audio is preprocessed using loudness normalization followed by neural denoising to ensure consistent audio quality for downstream models. For classification, we experiment with different model variants combining multilingual and Tamil-pretrained **Wav2Vec2** backbones with five temporal pooling strategies under frozen and partial fine-tuning settings. Our best configuration, i.e., learned attentive pooling with partial fine-tuning and a differentially trained MLP head, achieves a macro F1 of **0.79**, securing **1st place** with a margin of **0.26** points. For ASR, we propose two novel **dialect-conditioned Whisper** architectures{---}residual injection and cross-attention{---}that inject dialect embeddings from the trained classifier into the ASR pipeline. In addition, we evaluate a vanilla Whisper-Tamil fine-tuned baseline. The best model achieved a **WER of 0.90**, securing **8th place** in the shared task."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="s-etal-2026-wise">
<titleInfo>
<title>Wise@DravidianLangTech 2026: Dialect-Aware Tamil Speech Classification and Recognition via Cross-Pipeline Embedding Transfer</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ganesh</namePart>
<namePart type="given">Sundhar</namePart>
<namePart type="family">S</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hari</namePart>
<namePart type="given">Krishnan</namePart>
<namePart type="family">N</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gnanasabesan</namePart>
<namePart type="family">G</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Suriya</namePart>
<namePart type="family">KP</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jyothish</namePart>
<namePart type="given">Lal</namePart>
<namePart type="family">G</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bharathi</namePart>
<namePart type="given">Raja</namePart>
<namePart type="family">Chakravarthi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruba</namePart>
<namePart type="family">Priyadharshini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anand</namePart>
<namePart type="given">Kumar</namePart>
<namePart type="family">Madasamy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sajeetha</namePart>
<namePart type="family">Thavareesan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Saranya</namePart>
<namePart type="family">Rajiakodi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Subalalitha</namePart>
<namePart type="family">Navaneethakrishnan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhivya</namePart>
<namePart type="family">Chinnappa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Balasubramanian</namePart>
<namePart type="family">Palani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Malliga</namePart>
<namePart type="family">Subramanian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kogilavani</namePart>
<namePart type="family">Shanmugavadivel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ratnavel</namePart>
<namePart type="family">Rajalakshmi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Underline (Virtual)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-401-9</identifier>
</relatedItem>
<abstract>This paper presents the **Wise** system for the shared task on dialect-based speech processing in Tamil, addressing two subtasks: **(1) four-way dialect region classification** (Northern, Southern, Western, Central), and **(2) dialectal Tamil ASR**. All audio is preprocessed using loudness normalization followed by neural denoising to ensure consistent audio quality for downstream models. For classification, we experiment with different model variants combining multilingual and Tamil-pretrained **Wav2Vec2** backbones with five temporal pooling strategies under frozen and partial fine-tuning settings. Our best configuration, i.e., learned attentive pooling with partial fine-tuning and a differentially trained MLP head, achieves a macro F1 of **0.79**, securing **1st place** with a margin of **0.26** points. For ASR, we propose two novel **dialect-conditioned Whisper** architectures—residual injection and cross-attention—that inject dialect embeddings from the trained classifier into the ASR pipeline. In addition, we evaluate a vanilla Whisper-Tamil fine-tuned baseline. The best model achieved a **WER of 0.90**, securing **8th place** in the shared task.</abstract>
<identifier type="citekey">s-etal-2026-wise</identifier>
<location>
<url>https://aclanthology.org/2026.dravidianlangtech-1.71/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>447</start>
<end>452</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Wise@DravidianLangTech 2026: Dialect-Aware Tamil Speech Classification and Recognition via Cross-Pipeline Embedding Transfer
%A S, Ganesh Sundhar
%A N, Hari Krishnan
%A G, Gnanasabesan
%A KP, Suriya
%A G, Jyothish Lal
%Y Chakravarthi, Bharathi Raja
%Y Priyadharshini, Ruba
%Y Madasamy, Anand Kumar
%Y Thavareesan, Sajeetha
%Y Rajiakodi, Saranya
%Y Navaneethakrishnan, Subalalitha
%Y Chinnappa, Dhivya
%Y Palani, Balasubramanian
%Y Subramanian, Malliga
%Y Shanmugavadivel, Kogilavani
%Y Rajalakshmi, Ratnavel
%S Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages
%D 2026
%8 July
%I Association for Computational Linguistics
%C Underline (Virtual)
%@ 979-8-89176-401-9
%F s-etal-2026-wise
%X This paper presents the **Wise** system for the shared task on dialect-based speech processing in Tamil, addressing two subtasks: **(1) four-way dialect region classification** (Northern, Southern, Western, Central), and **(2) dialectal Tamil ASR**. All audio is preprocessed using loudness normalization followed by neural denoising to ensure consistent audio quality for downstream models. For classification, we experiment with different model variants combining multilingual and Tamil-pretrained **Wav2Vec2** backbones with five temporal pooling strategies under frozen and partial fine-tuning settings. Our best configuration, i.e., learned attentive pooling with partial fine-tuning and a differentially trained MLP head, achieves a macro F1 of **0.79**, securing **1st place** with a margin of **0.26** points. For ASR, we propose two novel **dialect-conditioned Whisper** architectures—residual injection and cross-attention—that inject dialect embeddings from the trained classifier into the ASR pipeline. In addition, we evaluate a vanilla Whisper-Tamil fine-tuned baseline. The best model achieved a **WER of 0.90**, securing **8th place** in the shared task.
%U https://aclanthology.org/2026.dravidianlangtech-1.71/
%P 447-452
Markdown (Informal)
[Wise@DravidianLangTech 2026: Dialect-Aware Tamil Speech Classification and Recognition via Cross-Pipeline Embedding Transfer](https://aclanthology.org/2026.dravidianlangtech-1.71/) (S et al., DravidianLangTech 2026)
ACL