AI-Based Inclusive Assistive And Learning Support Platform For Visually Impaired And Autism Students

Abstract

Inclusive education necessitates the development of technology solutions that can accommodate diverse learning abilities, especially those of visually impaired individuals and those with Autism Spectrum Disorder (ASD). This paper discusses the design and development of an Artificial Intelligence (AI) enabled platform for assistive and learning support in an inclusive education context. The proposed platform is divided into two main modules: a voice-enabled assistive module for visually impaired individuals, and a visual engagement analytics module for Autism Spectrum Disorder students. The assistive module is enabled by Automatic Speech Recognition (ASR), Optical Character Recognition (OCR), YOLO Object Detection, Text-to-Speech (TTS), and an SOS emergency feature. The learning analytics module incorporates eye-tracking data, feature engineering, random forest classification, SHAP explain ability, and argentic reasoning for analyzing student engagement. The publicly available eye-tracking data and synthetically generated data set have been used to test the classification of student engagement and behavioral pattern analysis. The proposed system architecture has adopted a scalable layered architecture for integrating perception modules, machine learning models, and explainable learning analytics modules. The proposed assistive computing and educational analytics framework can be useful for creating an inclusive learning environment.