Character of spoken dialogue systems is important not only for giving a positive impression of the system but also for gaining rapport from users. We have proposed a character expression model for spoken dialogue systems. The model expresses three character traits (extroversion, emotional instability, and politeness) of spoken dialogue systems by controlling spoken dialogue behaviors: utterance amount, backchannel, filler, and switching pause length. One major problem in training this model is that it is costly and time-consuming to collect many pair data of character traits and behaviors. To address this problem, semi-supervised learning is proposed based on a variational auto-encoder that exploits both the limited amount of labeled pair data and unlabeled corpus data. It was confirmed that the proposed model can express given characters more accurately than a baseline model with only supervised learning. We also implemented the character expression model in a spoken dialogue system for an autonomous android robot, and then conducted a subjective experiment with 75 university students to confirm the effectiveness of the character expression for specific dialogue scenarios. The results showed that expressing a character in accordance with the dialogue task by the proposed model improves the user's impression of the appropriateness in formal dialogue such as job interview.
We recorded dialogue videos where the first author and ERICA talked for 8 minutes in the two different dialogue scenarios of job interview and laboratory guide. In job interview, ERICA asked questions as the interviewer, and the first author responded as the interviewee. In the laboratory guide task, ERICA presented research topics of a laboratory and asked some questions to the visitor, and the first author listened to the explanation and answered questions as the visitor. For each task, we prepared two dialogue videos in two different conditions: proposed and baseline. In the proposed-method condition, the proposed character expression model was given a character that is regarded as appropriate for the corresponding dialogue task. In the baseline condition, ERICA spoke with a neutral character.
The robot expresses the emotional stable and polite character.
The robot expresses the neutral character.
The robot expresses the extrovert, emotional stable and casual character.
The robot expresses the neutral character.