Wise@DravidianLangTech 2026: Dialect-Aware Tamil Speech Classification and Recognition via Cross-Pipeline Embedding Transfer
Year: 2026
Type: conference-paper
Abstract: This paper presents the Wise system for the shared task on dialect-based speech processing in Tamil, addressing two subtasks: (1) fourway 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 experminet 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 1 st place (0.26point margin).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, and additionally evaluate a vanilla Whisper-Tamil finetuned baseline.The best model achieved a WER of 0.90, securing 8 th place.
Institution: Amrita Vishwa Vidyapeetham
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166797245