Designing and Evaluating an Adaptive Spoken Dialogue System

  • Authors:
  • Diane J. Litman;Shimei Pan

  • Affiliations:
  • University of Pittsburgh, Computer Science Department and LRDC, Pittsburgh, PA 15260 USA. E-mail: litman@cs.pitt.edu;IBM T.J. Watson Research Center, 30 Saw Mill River Road, Hawthorne, NY 10532 USA. E-mail: shimei@us.ibm.com

  • Venue:
  • User Modeling and User-Adapted Interaction
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

Spoken dialogue system performance can vary widely for different users, as well for the same user during different dialogues. This paper presents the design and evaluation of an adaptive version of TOOT, a spoken dialogue system for retrieving online train schedules. Based on rules learned from a set of training dialogues, adaptive TOOT constructs a user model representing whether the user is having speech recognition problems as a particular dialogue progresses. Adaptive TOOT then automatically adapts its dialogue strategies based on this dynamically changing user model. An empirical evaluation of the system demonstrates the utility of the approach.