Second International Conference on Spoken Language Processing (ICSLP'92)

Banff, Alberta, Canada
October 13-16, 1992

A Flexible Multimodal Dialogue Architecture Independent of the Application

Philippe Morin (1,2), Jean-Claude Junqua (1,3), Jean-Marie Pierrel (2)

(1) Speech Technology Laboratory, Division of Panasonic Technologies, Inc., Santa Barbara, California, USA
(2) CRIN-CNRS/INRIA Lorraine, Vandoeuvre les Nancy, France
(3) Central Research Laboratories, Matsushita, Japan

In this paper, we present the general architecture of the multimodal dialogue system PARTNER which is designed for goal-oriented applications using speech input and output. Organized around three major components: an I/O manager; a Dialogue manager, and an Application manager, its architecture permits easy implementation of new applications, and new modalities as well as different languages. A common meaning representation language is used as a vehicle for internal communications. Our objective is to design and implement a Man-Machine dialogue interface that is generic and efficient while emphasizing its human-likeness. Currently, the proposed architecture is being validated for several real-world applications.

Full Paper

Bibliographic reference.  Morin, Philippe / Junqua, Jean-Claude / Pierrel, Jean-Marie (1992): "A flexible multimodal dialogue architecture independent of the application", In ICSLP-1992, 939-942.