Students face a universal challenge during lectures where they cannot simultaneously listen attentively while taking comprehensive notes, often missing important concepts and struggling later to review hours of recordings or incomplete notes for exam preparation. This project will create an AI-powered lecture companion that provides real-time transcription synchronized with slides, automatically identifies and highlights key concepts when professors emphasize points, organizes content into searchable hierarchical structures with timestamps, and generates study aids like flashcards and summaries. The research will investigate how AI assistance affects student attention and learning outcomes, which interface designs help students effectively use AI-generated content, and how to implement universal design principles that make the system accessible to students with disabilities while benefiting all learners. The system will directly address the difficulty that human attention and writing speed are limited resources, forcing students to choose between actively listening to understand concepts versus mechanically transcribing information. This work will combine natural language processing, information retrieval, and educational interface design to contribute evidence-based guidelines for educational AI tools and produce a working prototype that could help thousands of Dal students.