Resolution of Lexical Ambiguities in Spoken Dialogue System
Berenike Loos, Robert Porzel · Annual Meeting of the Special Interest Group on Discourse and Dialogue · 2004
The development of conversational multidomain spoken dialogue systems poses new challenges for the reliable processing of less restricted user utterances. Unlike in controlled and restricted dialogue systems a simple oneto-one mapping from words to meanings is no longer feasible here. In this paper two different approaches to the resolution of lexical ambiguities are applied to a multi-domain corpus of speech recognition output produced from spontaneous utterances in a spoken dialogue system. The resulting evaluations show that all approaches yield significant gains over the majority class baseline performance of .68, i.e. fmeasures of .79 for the knowledge-driven approach and .86 for the supervised learning approach.