Skip to main content Skip to navigation

AI for Accessible Education: Personalized Audio-Based Learning for Blind Students

Project Overview

The document explores the application of generative AI in education, focusing on 'Audemy', an innovative audio-based learning platform specifically designed for blind and visually impaired (BVI) students. This platform leverages adaptive learning techniques to create personalized and engaging educational experiences, tailoring content to meet individual learning needs. Developed with input from educators, Audemy incorporates features such as real-time feedback and conversational AI to enhance the learning process. Additionally, the document highlights the importance of addressing ethical concerns, particularly regarding data privacy and security, ensuring that the platform not only supports educational advancement but also respects the rights of its users. Overall, the findings suggest that generative AI can significantly improve accessibility and personalization in education, making it a valuable tool for fostering inclusive learning environments.

Key Applications

Audemy, an AI-powered audio-based learning platform

Context: For blind and visually impaired (BVI) students in traditional educational settings

Implementation: Developed iteratively with feedback from over 20 educators specializing in accessibility. Features include adaptive learning, real-time feedback, and conversational AI.

Outcomes: Increased engagement and personalized learning experiences for over 2,000 BVI students, addressing their unique educational needs.

Challenges: Limited ability to interpret complex emotional states and make context-sensitive decisions based on audio cues.

Implementation Barriers

Technical Barrier

Current limitations in AI's ability to interpret emotional states through audio cues.

Proposed Solutions: Future research can focus on incorporating emotional assessment techniques into AI systems.

Ethical Barrier

Concerns about data privacy and security, especially when handling sensitive information from minors.

Proposed Solutions: End-to-end encryption, clear guidelines on data usage, and separation of voice inputs from personally identifiable information.

Project Team

Crystal Yang

Researcher

Paul Taele

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Crystal Yang, Paul Taele

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies