부산대학교 도서관

People who are visually impaired or blind may find it difficult to comprehend what is going on around them. This project proposes an AI-based application that when the user gives a voice command, the application recognizes the voice command and takes the visual data (image) with an externally connected camera from the surroundings of blind or visually impaired people and based on the voice command and it will be sent to corresponding machine learning model API and then that model may generate the captions or textual description in natural language(English) or different kind of objects present in the image and finally convert the same into speech in natural language using Text to Speech API so that user can hear it. Now by using this AI application, blind and visually impaired people who are mostly dependent on volunteers or any assistive materials in moving around in their surroundings can be less dependent and also move around freely.