JCSE, vol. 2, no. 1, pp.1-25, 2008

DOI:

Using Utterance and Semantic Level Confidence for Interactive Spoken Dialog Clarification

Sangkeun Jung, Cheongjae Lee, Gary Geunbae Lee
Computer Science and Engineering, Pohang University of Science and Technology

Abstract: Spoken dialog tasks incur many errors including speech recognition errors, understandingerrors, and even dialog management errors. These errors create a big gap between the user’sintention and the system’s understanding, which eventually results in a misinterpretation. Tofill in the gap, people in human-to-human dialogs try to clarify the major causes of themisunderstanding to selectively correct them. This paper presents a method of clarificationtechniques to human-to-machine spoken dialog systems. We viewed the clarification dialog as atwo-step problem - Belief confirmation and Clarification strategy establishment. To confirmthe belief, we organized the clarification process into three systematic phases. In the beliefconfirmation phase, we consider the overall dialog system’s processes including speechrecognition, language understanding and semantic slot and value pairs for clarification dialogmanagement. A clarification expert is developed for establishing clarification dialog strategy.In addition, we proposed a new design of plugging clarification dialog module in a given expertbased dialog system. The experiment results demonstrate that the error verifiers effectivelycatch the word and utterance-level semantic errors a

Keyword: No keyword

Full Paper:   286 Downloads, 4558 View