Using prosodic clues to decide when to produce back-channel utterances

Back-channel feedback is required in order to build spoken dialog systems that are responsive. This paper reports a model of back-channel feedback in Japanese dialog. It turns out that a low pitch region is a good clue that the speaker is ready for back-channel feedback. A rule based on this fact matches corpus data on respondents' production of back-channel feedback. A system based on this rule meets the expectations of live speakers, sometimes well enough to fool them into thinking they are conversing with a human.