A Maximum Entropy model for modelling context in a spoken dialogue system

Rob Koeling
RijksUniversiteit Groningen, department of Alfa-Informatica
koeling@let.rug.nl

Within the NWO priority program OVIS a dialogue system is developed with which information about the railway schedule can be obtained via the telephone in spoken Dutch. A speech recogniser analyses the user utterance and returns a (possibly big) number of hypotheses to the natural language processing component. It is the task of the NLP component to find the best hypothesis. To reach this goal we try to exploit as many information sources as possible (e.g. linguistic, accoustic and statistical (n-gram)) An extra source of information that is investigated is the use of context. In this talk I discuss an ongoing experiment to model contextual information by means of a statistical model (Maximum Entropy). I try to model dependencies between words in the user utterance and information like the type of or the words included in the corresponding system question. Maximum Entropy models are a popular means today for combining information from different sources.