Final Year Project

Project Title

YAARAN

Description

My project focuses on developing an AI-based speech translation model that converts Burushaski audio into English text. Burushaski is primarily a spoken language with very limited written resources and no publicly available digital datasets and is increasingly at risk as younger generations shift toward Urdu and English. The project aims to support language preservation while improving accessibility through AI technologies. Since no dataset currently exists, our team is collecting, curating, and annotating original audio data through multiple community-based data collection methodologies. The curated dataset will be used to train and evaluate the translation model. Beyond Burushaski, YAARAN also aims to establish a framework that could enable the development of AI translation systems for other low-resource spoken languages in the future.