Enhancing Primary Students' Oral Proficiency With AIPowered Chatbots: Introducing Learn
DOI:
https://doi.org/10.64252/hj2dkw44Keywords:
AI in education, automated speech evaluation, chatbot learning, Mother Tongue Language, oral proficiency, primary educationAbstract
This study presents LEARN (Language automated Evaluation by generating Answers/Questions from caRtooNs), an innovative AI-powered system designed to improve oral proficiency among Primary 1 and 2 students learning their Mother Tongue Languages (MTL) in Singapore. Jointly developed by the National Institute of Education (NIE) – Nanyang Technological University (NTU), and the Singapore Institute of Technology (SIT), LEARN engages students in interactive conversations based on picture-based prompts. The project aims to enhance the home-based learning environment by leveraging conversational AI chatbots that simulate engaging picture-based dialogues. Incorporating advanced technologies such as visual dialogue engines, automated speech recognition, and gamification, LEARN targets a 20% increase in oral proficiency over two years. Data collected from 659 students across ten primary schools were used to train LEARN. Preliminary findings and achievementssuggest strong feasibility, linguistic impact, and scalable potential. This paper discusses LEARN as anedtech solution to improve MTL oral proficiency.