When the target language must mark a distinction the source does not, a translation system must still pick a value. English they came becomes one Tamil verb if those who came are people and another if they are not; English does not say which. We ask what automatic evaluation charges when the choice is wrong, or never made. We release TamilLingBench, an English→Tamil challenge set for three verb-agreement distinctions, checked by a morphological analyser: rationality (திணை tiṇai), gender, and number as a control. We pre-registered a prediction that neural metrics would not notice a single wrong morpheme. It is wrong for the neural reference-based metrics: COMET-22 charges 8.52× the score difference that human raters reliably notice. On the systems’ own errors, both reference-free metrics rank the wrong form first more often than not, and a legitimate Tamil form that leaves the distinction out is not charged at all. Under reference-free evaluation, dropping the distinction is free; reference-based metrics do charge for it. xCOMET finds the error’s region but not the morpheme. Inside one model, activation patching finds the same asymmetry: a distinction English marks is settled by the middle of the network, one it leaves unmarked only near the output.

This study introduces a novel multilingualmodel designed to effectively address the challenges of detecting abusive content in low resource, code-mixed languages, where limiteddata availability and the interplay of mixed languages, leading to complex linguistic phenomena, create significant hurdles in developingrobust machine learning models. By leveraging transfer learning techniques and employingmulti-head attention mechanisms, our modeldemonstrates impressive performance in detecting abusive content in both Tamil and Malayalam datasets. On the Tamil dataset, our teamachieved a macro F1 score of 0.7864, whilefor the Malayalam dataset, a macro F1 score of0.7058 was attained. These results highlight theeffectiveness of our multilingual approach, delivering strong performance in Tamil and competitive results in Malayalam.

This paper introduces EmoTa, the first emotional speech dataset in Tamil, designed to reflect the linguistic diversity of Sri Lankan Tamil speakers. EmoTa comprises 936 recorded utterances from 22 native Tamil speakers (11 male, 11 female), each articulating 19 semantically neutral sentences across five primary emotions: anger, happiness, sadness, fear, and neutrality. To ensure quality, inter-annotator agreement was assessed using Fleiss’ Kappa, resulting in a substantial agreement score of 0.74. Initial evaluations using machine learning models, including XGBoost and Random Forest, yielded a high F1-score of 0.91 and 0.90 for emotion classification tasks. By releasing EmoTa, we aim to encourage further exploration of Tamil language processing and the development of innovative models for Tamil Speech Emotion Recognition.