Home
 

 versão em português

 

Home

Overview

Research

Development

 

Conversão Texto-Fala 

Echo Cancellation 

Other Services

Registration

Webmail

Site map

Contact

Text-To-Speech Conversion System

In the speech synthesis area, since 1998, LINSE research group has developed some versions of text-to-speech (TTS) systems. This research project has been financially supported by Dígitro Tecnologia Ltda., the Brazilian National Research Council for Scientific and Technological Development (CNPq), and the Studies and Projects Funding Body (FINEP).

The approach used in these systems is the concatenative synthesis. Our first TTS system consisted of a group of fixed units. The repertoire of units for this system comprised especially triphones and some poliphones (see, in FIRST VERSION, some sentences from our first text-to-speech system).

Nowadays, our TTS considers only one phoneme or even larger units such as syllables, words, phrases and short sentences (see, in NEW VERSION, sentences from our current text-to-speech system).

  First Version Samples

  New Version Samples

  Publications in the Area

LINSE has the right to change the content and format of this website whenever necessary.

This site has been totally developed by LINSE. All rights reserved @2007.