What are Multimodalities Made of? Modeling Output in a Multimodal Dialogue System

Christian Elting

In order to produce coherent multimodal output a presentation planner in a multimodal dialogue system must have a notion of the types of the multimodalities, which are currently present in the system. More specifically the planner needs information about the multimodal properties and rendering capabilities of the multimodalities. Therefore it is necessary to define an output multimodality model that can properly describe the available renderers in sufficient detail and on the other hand keep a level of abstraction that enables the presentation planner to support a large set of different renderer types. In this paper we present our approach for such a multimodality model.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.