Indonesian Text to Audio Visual Speech with Animated Talking Head

Abstract

Most people have to connect to the computer system for accommodating their daily routines. For this reason, a supporting system must be available so that a natural communication between users and the computer system can run smoothly and naturally. Speech is a means of communication in which a visual speech information must be integrated in order to improve and to sharpen up the quality of its naturalness. The most ideal way to enhance the quality of human and computer interaction is by applying the audio visual speech mode. Text To Audio Visual Speech (TTAVS) is a computer system able to transform text input into speech message and to enrich synthesized visual speech for being delivered as a unit of message to the user simultaneously. This paper is addressed to the development of TTAVS in Indonesian language. Mean Opinon Score (MOS) test is used to evaluate the performance of the system. Indonesian natives, among them some lectures of Indonesian, have been selected to participate in this test. The average outcome based on the used MOS test indicates a promising result.