Image to speech synthesizer with reference to assamese numerals

Abstract

In this research work we have shown the methodology for converting printed Assamese numerals to its corresponding utterance. We have implemented as an initial effort which will read only four digit numerals. We are using Image processing techniques to convert an image of Assamese numerals into textual/digital form. In the second phase the numerals will be pronounced as a number by Google speaker. In this system, images are stored in a dataset and then inputted data is compared with the dataset image using template matching technique. After recognition of the text output will be displayed as a speech waveform. This work has many applications in today’s digital world.