Implementation of Telugu Speech Synthesis System

Speech synthesis is the computer generated human voice. It is also known as a text-to-speech system which converts text information into speech. Speech synthesis systems are often called text-to-speech (TTS) systems about their ability to convert text into speech. A TTS synthesis system converts written orthographic text into corresponding artificial speech signals. In multi-lingual cultural settings, listeners expect a high-quality TTS synthesis system to read the text in a polyglot manner, i.e., in such a way that the origin of the inclusions is heard, by using correct language-specific pronunciation and prosody. Multilingual TTS synthesis systems are unable to convert such mixed-lingual texts into polyglot speech signals correctly. This paper presents a report on the work done to construct a polyglot TTS synthesis system which able to convert mixed lingual text into polyglot speech signals. Here mixed lingual text analyzer is proposed from a combination of monolingual text analyzers and compare both mixed-lingual and multi-lingual systems.