@inproceedings{abhinav-etal-2026-dlrg,
    title = "{DLRG}@{D}ravidian{L}ang{T}ech 2026: Dual-Purpose Whisper Adaptation for {T}amil Dialect Identification and Dialectal Speech Recognition",
    author = "Abhinav, Gulisetty  and
      Nanda, Tanisha  and
      R, Ramesh Kannan  and
      Rajalakshmi, Ratnavel",
    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://preview.aclanthology.org/ingest-acl-workshops/2026.dravidianlangtech-1.33/",
    pages = "232--236",
    ISBN = "979-8-89176-401-9",
    abstract = "This paper describes our system developed for the shared task on Dialect Based Speech Recognition and Classification in Tamil at DravidianLangTech@ACL 2026. We participated in both Subtask{~}1 (Dialect Identification) and Subtask{~}2 (Dialectal ASR). Our approach leverages a single Tamil-adapted Whisper Medium model as a unified foundation for both tasks. For dialect classification, we have used the Whisper encoder as a feature extractor by discarding the decoder, applying mean pooling over the temporal dimension, and fine-tuning the full encoder with a lightweight classification head, achieving 73.4{\%} accuracy on the test set. For dialectal ASR, we apply Low-Rank Adaptation (LoRA) to the full encoder-decoder architecture with SpecAugment-based data augmentation, achieving a Word Error Rate (WER) of 0.55 on the test set. Our experiments reveal that unfreezing the pre-trained encoder is critical for dialect discrimination, boosting accuracy from 52.78{\%} (frozen) to 73.4{\%} (unfrozen). The code is publicly available at https://github.com/DLRG-VIT/DravidianLangTech2026"
}