Audio reading system for blind persons, testing and evaluation.

According to the World Health Organization, there are approximately 285 million blind people around the world. Those people faced challenges in their life, especially when reading a book. This paper is aims to test and evaluation the audio reading system (which is use to help the blind person to read the book) with new dataset which consist of 161 image, 8 font size, mono and color image, and different views. The system is test and discussion from the aspects: accurate, Font size and type, and more reliable. According to the results (on our dataset (image type, 161 images)), the characteristics of the system is efficient and accurate real time system with cloud computing and the accuracy increase while the font size increase, whereas the accuracy of font size 30 is 100% and of 12 is 80%. [ABSTRACT FROM AUTHOR]