This paper considers three dialogue control strategies to relieve speech recognition errors. These are the prompt to speak again (basic strategy), the direct confirmation and the indirect confirmation. The purpose of modeling the dialog control strategies is to estimate two quantities Pac and N, given the performance of the speech recognizer used in a dialogue system. Pac is the probability that information included in user's utterance is conveyed to the system correctly, and N is the average number of turns taken between the user and the system until terminating subdialogue on user's first utterance. The analysis has proven that the direct confirmation can increase Pac and the indirect confirmation can reduce N in comparison with the basic strategy. Since the mathematical analysis described in this paper can easily be extended to more sophisticated strategies, it will contribute to the quantitative design of a dialogue control strategy, give the performance of a speech recognizer.
Bibliographic reference. Niimi, Y. / Kobayashi, Y. (1995): "Modeling dialogue control strategies to relieve speech recognition errors", In EUROSPEECH-1995, 1177-1180.