Publication Abstract

Tummala, D., Seneff, S., Paul, D. B., Weinstein, C. J., and Yang, D., CCLINC: System Architecture and Concept Demonstration of Speech- to-Speech Translation for Limited-Domain Multilingual Applications, Proceedings of the 1995 ARPA Spoken Language Technology Workshop, Austin, TX, January 1995, pp. 227-232.

Abstract

This paper describes CCLINC, a system architecture and concept demonstration for automatic speech-to-speech translation for limited-domain multilingual applications. The primary target application is the coalition battle management environment. CCLINC utilizes a Common Coalition Language (CCL) as a military interlingua. CCLINC is a speaker-independent system that translates spoken utterances in English into French or Korean. The current system has a vocabulary of about 700 words. The system architecture for CCLINC consists of a modular, multilingual structure including speech recognition, language understanding, language generation, and speech synthesis in each language. A key new feature of the system is the tight coupling of the speech recognition and language understanding modules. We summarize the architectures of the component systems and the interfaces between them, and present our preliminary performance results.