Automatic Speech Recognition (ASR) systems for morphologically complex languages like Urdu, Persian, and Arabic face unique challenges due to the intricacies of Perso-Arabic scripts. Conventional data processing methods often fall short in effectively handling these languages’ phonetic and morphological nuances. This paper introduces a unified data processing pipeline tailored specifically for Perso-Arabic languages, addressing the complexities inherent in these scripts. The proposed pipeline encompasses comprehensive steps for data cleaning, tokenization, and phonemization, each of which has been meticulously evaluated and validated by expert linguists. Through expert-driven refinements, our pipeline presents a robust foundation for advancing ASR performance across Perso-Arabic languages, supporting the development of more accurate and linguistically informed multilingual ASR systems in future.

This paper presents the submission of IIITH-BUT to the IWSLT 2025 shared task on speech translation for the low-resource Bhojpuri-Hindi language pair. We explored the impact of hyperparameter optimisation and data augmentation techniques on the performance of the SeamlessM4T model fine-tuned for this specific task. We systematically investigated a range of hyperparameters including learning rate schedules, number of update steps, warm-up steps, label smoothing, and batch sizes; and report their effect on translation quality. To address data scarcity, we applied speed perturbation and SpecAugment and studied their effect on translation quality. We also examined the use of cross-lingual signal through joint training with Marathi and Bhojpuri speech data. Our experiments reveal that careful selection of hyperparameters and the application of simple yet effective augmentation techniques significantly improve performance in low-resource settings. We also analysed the translation hypotheses to understand various kinds of errors that impacted the translation quality in terms of BLEU