This paper describes the use of a simulation method to assess speech interactive dialogues. Two alternative telephone banking dialogues were modelled, both of which were assumed to produce the same eventual task output, but which used sets of prompts which would impose differing levels of constraint over user utterances. It was found that although increasing constraint would improve recognition accuracy in the task, this was not enough to counteract the increased length of interaction produced by the more constraining prompts. Thus the less constraining dialogue was found to be faster despite its increased recognition error rate compared to the more constrained dialogue. Level of constraint is put forward in this paper as a central issue in speech dialogue design, both because of its supposed effect on user satisfaction, and because of its implications for system performance.
Cite as: Hone, K.S., Baber, C. (1995) Using a simulation method to predict the transaction time effects of applying alternative levels of constraint to user utterances within speech interactive dialogues. Proc. ESCA Workshop on Spoken Dialogue Systems, 209-212
@inproceedings{hone95_sds,
author={K. S. Hone and C. Baber},
title={{Using a simulation method to predict the transaction time effects of applying alternative levels of constraint to user utterances within speech interactive dialogues}},
year=1995,
booktitle={Proc. ESCA Workshop on Spoken Dialogue Systems},
pages={209--212}
}