An architecture for voice dialog systems based on prolog-style theorem proving

  • Authors:
  • Ronnie W. Smith;Alan W. Biermann;D. Richard Hipp

  • Affiliations:
  • East Carolina University;Duke University;Hipp, Wyrick & Company, Inc.

  • Venue:
  • Computational Linguistics
  • Year:
  • 1995

Quantified Score

Hi-index 0.00

Visualization

Abstract

A pragmatic architecture for voice dialog machines aimed at the equipment repair problem has been implemented. This architecture exhibits a number of behaviors required for efficient human-machine dialog. These behaviors include:(1) problem solving to achieve a target goal(2) the ability to carry out subdialogs to achieve appropriate subgoals and to pass control arbitrarily from one subdialog to another(3) the use of a user model to enable useful verbal exchanges and to inhibit unnecessary ones(4) the ability to change initiative from strongly computer controlled to strongly user controlled or to some level in between(5) the ability to use context dependent expectations to correct speech recognition and track user movement to new subdialogs.The paper gives a description of the dialog theory, presents examples of its capabilities, and includes a detailed trace of one of those examples showing all significant mechanisms. The paper gives performance data for a series of 141 problem-solving dialogs carried out with human subjects.