Gujarati Text-To-Speech (TTS) conversion system using Histogram and Edge detection method

Text-To-Speech (TTS) takes raw text as an input and converts it into equivalent sound as output. TTS system quality can be measured by two parameters: naturalness and intelligibility. It is very difficult to achieve both the parameters simultaneously in the single system. Thus, there is a great scope of research in this topic. Scientists are still working to give expression and feelings to this type of system but not successfully achieved the results. Blind and illiterate people can use widely use this TTS system for gaining knowledge from book. Public places like airport, railway stations can use this system for repetitive announcements. Input for this system can be typed or scanned text. We can hear our own voice in the output sound by superimposing our voice frequency over the recorded frequency. This paper begins with the fundamental concepts of speech synthesis, so that nontechnical person can also contribute in research.