Recent Trends in Text to Speech Synthesis of Indian Languages

Author Name(s): Sarang L. Joshi, Vinayak K. Bairagi
Author Email:


A Text To Speech (TTS) synthesizer is a computer application capable of converting arbitrary input text into speech. This conversion broadly involves two steps, namely, text processing and speech synthesis. Text processing converts the entered text to a sequence of synthesis units, while speech synthesis is the generation of an acoustic wave form corresponding to each of these units. Naturalness and intelligibility are the most important qualities expected from a TTS system. In this paper we aim to provide an overview of various techniques for text to speech synthesis, discuss their characteristics, summarize and compares advantages and drawbacks. We have listed various Text-to-Speech synthesis frameworks developed and implemented at different Indian institutes.


Speech is one of the premier forms of everyday life communication among humans. Most of the information in digital world is accessible only to those few people who can read or understand a particular language. Computers communicating with the common man in the language he understands and convenient for him can make the digital content reach to the masses by facilitating the exchange of information across different people speaking different languages. For visually impaired, speech impaired, educationally under privileged and the rural communities, text to speech (TTS) conversion can reduce the trouble in human interaction with computers and make it as simple as ABC (Kishore et al.,2003; Gaikwad et al., 2014).


The linguistic structure of each language is different hence synthesis rules are different for all languages. Though the research work has extended to cover few more languages, still many languages are left to be covered. A syllable can be the best unit for Indian language speech synthesis as Indian languages are syllable centred i.e. pronunciation is mainly syllable based. Prosody plays a crucial role in making speech sound natural. As a field, study of the prosody of Indian languages is still in its infancy. There is dire need of working on common platforms / fonts / standards / procedures / software tools. More resources should be made available electronically with fewer restrictions.

204 total views, 1 views today

Download PDF File

About the author: admin