Probabilistic, Multi-staged Interpretation of Spoken Utterances

Abstract. We describe Scusi?, a multi-stage, spoken language interpretation mechanism designed to be part of a robot-mounted dialogue system. Scusi?’s interpretation process maps spoken utterances to conceptual graphs, and the nodes in these graphs to concepts in the world. Maximum posterior probability is used to rank the (partial) interpretations produced at each stage of this process.

• Type: Conference
• Year: 2006
• Where: CIMCA
• Authors: Ingrid Zukerman, Michael Niemann, Sarah George
• Keywords: CIMCA 2006, Interpretation Process Maps, Maximum Posterior Probability, Spoken Language Interpretation