Smart Vision: Developing an OCR-Based Speech Synthesis System with LabVIEW

Abstract

In today’s world we are surrounded by one thing and that is data. This data needs to be recorded for various purposes and to do this manually is a huge task. Text is one of the forms of data and to record this text in computers is very difficult since typing all the text is time consuming and therefore inefficient. Hence, we can make this process more efficient by using Optical Character Recognition (OCR). OCR can help in conversion of this text data to speech signal. This efficient conversion to speech opens more doors for further applications. One of which is for blind and visually impaired, by which they can easily comprehend the text data without relying on third party for help. This paper aims to show the development of the OCR based speech synthesis system that is cost effective and easy to understand. The OCR converts the image to text which it then converts to speech. After the use of OCR, when the image is converted into speech, speech libraries present, such as in Microsoft SDK, are used for the conversion of text to speech. In this manner, the user, after scanning the image, is able to receive the information in form of speech. OCR training can be done in multiple languages and multiple fonts as well as handwritten font, according to the convenience of the user. Thus, software technology using Lab view is used for easy reception of written data using speech. This system has applications in finance industry, medical, home automation, etc. National Instrument’s LabVIEW has been used to develop OCR based speech synthesis system