A) Client- A start up company - Distributed Speech Recognition System

Venus is distributed speech recognition system. It operates in two modes, Link Redirection mode and Content Information mode. In Link Redirection mode it takes speech as input; uses Voice Recognition Engine to recognize voice and converts it to text; uses its database information to map recognized text to proper URL string. With Content Information mode rather than returning URL strings it returns HTML information. In this mode it processes HTML data to be returned like convert it to WML or XML.

A Multi-model prototype Browser was used to test the capability of the server. This prototype browser takes speech as input; processes speech and sends the encoded speech to Venus server using HTTP POST method. It also retrieves the result that is either list of URL strings or list of HTML pages and displays the result.

Environment: JDK 1.3, Weblogic 5.10, Apache XML Parser for Java (Xerces), TogetherJ (UML tool), Window 2000 server and Red Hat Linux 7.0