Effective human-computer cooperative spoken dialogue: the AGS demonstrator

Abstract

This paper describes a spoken dialogue system based on a generic specication of a cooperative communicating rational agent. We present some theoretical and practical apects of the overall approach, along with the speech-specic and natural language related issues raised by the eective implementation of the system. An account is also given of an evaluation of the system with naive users on the task of voice services directory (AGS) inquiry.

On the use of finite state transducers for semantic interpretation

“…Experiments are carried out on a dialogue corpus provided by France Telecom R&D and collected for a tourism telephone service (Sadek et al, 1996). Two corpora, extracted from this dialogue corpus, are used in this study:…”

“…The dialogue corpus used for training and evaluating these algorithms is made of dialogue traces of the France Telecom R&D spoken dialogue system PlanResto (Sadek et al, 1996) which is a booking restaurant application. All the examples presented here are related to this application.…”

Multi-Agent Programming

“…To be used in such a context, the rational unit requires natural language processing components [198,165] that bridge the gap between the human language and the internal semantic knowledge representation in terms of communicative acts with semantic contents expressed in a powerful language (a first-order modal language): ARCOL (see section 9.4). It can be used as a regular communicating agent in a multi-agent infrastructure.…”

Construct Algebra

“…No explicit states and transitions need to be defined using this framework (Bennacef et al, 1996;Meng and et. al., 1996;Sadek et al, 1996). A change to the dialog does not require a change to the dialog manager, but more simply, a change to the inheritance hierarchy.…”