Abstract

This study explores computational approaches to analyzing dialectal variations in the Odia language, with a focus on native language identification. Odia, a prominent language spoken in India, exhibits significant regional dialectal diversity, which can pose challenges in automatic language processing tasks. This research investigates various computational techniques for identifying native Odia speakers by analyzing dialectal features, such as phonetic, lexical, and syntactic variations, across different regions. We employ machine learning models, including supervised and unsupervised learning, to classify and distinguish between different dialects of Odia. By leveraging natural language processing (NLP) tools and a large corpus of dialectal data, the study aims to enhance the accuracy of language identification systems, which are critical for applications in speech recognition, language modeling, and automated translation. The findings highlight the potential of computational methods to capture the subtle nuances of dialectal differences and improve the performance of language processing systems for less-resourced languages like Odia.