Human Computer Interaction via Spoken Dialogue


James Allen

Institute for Human and Machine Cognition


A conversational agent is a system that can engage in natural language conversation in order to further its goals. To build such a system requires a concerted effort to bring work in speech recognition, natural language understanding, dialog modeling, planning, knowledge representation and reasoning into a single coherent system. I will describe our ongoing work on the TRIPS system, a conversational agent that can support unconstrained dialogue in order to assist the user in problem solving tasks. Significant effort has been made to make the system robust, so that it can perform well even in the face of inevitable speech recognition errors, and can continue the dialogue in a natural way under any circumstances. This talk will discuss the overall approach, and highlight various aspects of the system that are the foci of our work.