Context modeling for language and speech generation

Some of the most important issues in the design of a dialogue system involve the modeling of linguistic context. The paper highlights a number of these issues, focusing an the language and speech generation components of such systems, and discusses their implications for the way in which context has to be modeled in a spoken dialogue system. We compare the `dedicated' context models that have been proposed in theoretical and computational linguistics with the more general models proposed in artificial intelligence. Our main examples of a dedicated context model is the context model of the Dial Your Disc (DYD) music information system (Van Deemter and Odijk, 1997) and the better-known discourse representation theory of which this model is a variant. Our main example of a `general' context model is provided by the so-called `1st' formalism (McCarthy, 1993