Third International Conference on Spoken Language Processing (ICSLP 94)
This paper describes effects on utterances caused by knowledge about the hearer. Our study focuses on the difference of expressions in the dialogue of two groups. Two groups of six and seven subjects were asked to obtain traffic information and choose a route using our car navigation system. Subjects in the system responding (SR) group were led to believe they were talking to a machine. Subjects in the human responding (HR) group were told they were talking to a human. Both groups of subjects actually talked to the human operators. Our car navigation system displayed a map and proposed three routes, one at a time, to the subject. Operators responded to subjects by selecting sentences in the system which were shown as a synthetic voice. Every subjects took part in three sessions within two weeks. Utterances of the subjects in both groups declined to be more concise expression as the session proceeded. However, subjects in the HR group used more various expressions, with respect to "intonation" and "accent" than the SR group. We observed more natural expressions as "filler" in the HR group. And we found that subjects in HR group tended to make shorter pauses in their utterances and more often to utter ill-formed sentences than in the SR group. These differences in expressions between two groups were seen through the whole sessions.
Bibliographic reference. Nakazato, Shu / Shirai, Katsuhiko (1994): "Effects on utterances caused by knowledge on the hearer", In ICSLP-1994, 975-978.