AI-assisted design of microlearning media for basic japanese grammar

Abstract

The objective of this study is to evaluate the feasibility of AI-assisted microlearning-based instructional videos for teaching Basic Japanese Grammar 1 (Bunpou I) at Universitas Negeri Jakarta. This research addresses the persistent challenge of limited classroom time and the lack of accessible, engaging self-study resources for first-semester Japanese learners, who often struggle with unfamiliar scripts and complex grammar structures. Recognizing the learning preferences of Generation Z—who favor short, visually rich, and technology-driven content—this study integrates Artificial Intelligence tools into the instructional design process to develop concise, focused explainer videos. Employing a Research and Development (R&D) methodology based on the PPE model (Planning, Production, Evaluation), the study involved syllabus analysis, AI-assisted media production, and expert validation, followed by a preliminary student trial. The instructional video received high feasibility scores from both media (3.62) and content experts (3.60), and student feedback indicated improved comprehension and engagement. The novelty of this study lies in its systematic integration of AI tools—such as Gemini AI, Canva Magic Media, and D-ID AI—for storyboard generation, visual content creation, and narration. The findings suggest that this AI-integrated microlearning model offers a scalable and effective instructional design framework for foreign language education, particularly for beginner learners in digital learning environments.