This paper presents a prototype to help visually impaired persons in reading printed learning materials using Raspberry PI. Tesseract an open source optical character recognition technique is used extract texts in printed images and converted to audio output using text-to-speech conversion software. Prototype is experimented using printed text pages with various font sizes and line spacing as test cases. Results show that the prototype is better in converting printed texts to speech. However quality of image, font size, and line space affects performance of prototype in converting printed texts to speech. © 2020 Elsevier B.V., All rights reserved.