Confidence-based adaptivity in response generation for a spoken dialogue system

Abstract

The paper addresses the issue of how to increase adaptivity in response generation for a spoken dialogue system. Realization strategies for dialogue responses depend on communicative confidence levels and interaction management goals. We first describe a Java/XML-based generator which produces different realizations of system responses based on agendas specified by the dialogue manager. We then discuss how greater adaptivity can be achieved by using a set of distinct generator agents, each of which is specialized in its realization strategy (e.g. highly elliptical or highly explicit). This allows a simpler design of each generator agent, while increasing the overall system adaptivity to meet the requirements for flexible cooperation in incremental and immediate interactive situations.