Text-To-Speech (TTS) Conversion System for Gujarati Language

Abstract

Text-to-speech (TTS) conversion system enables user to enter text in Gujarati language and as an output it generates the equivalent sound. This type of system will be greatly useful for illiterate and vision-impaired people to hear and understand the content. TTS systems are still suffering from the problem of producing emotional speech like that of human being. Scientists are trying to give emotions and feelings to it. This shows that research work can be done to enhance the efficiency of TTS system. There are TTS systems that are under development for many languages other than Gujarati language. So, we are proposing the TTS system, which will work on Gujarati language. The input of our desired system is typed or scanned Gujarati text and equivalent Gujarati speech with smooth flow will be generated as an output. We are trying to add a new feature through which we can hear our own voice by superimposing our own voice frequency on the pre-recorded synthesized speech, so that we can listen to any text in our own voice. This paper starts with the introduction to the fundamental concepts of TTS synthesis. So, it will be useful for the readers who are less familiar in this area of research.