Paired Conversational Agents for Easy-to-Understand Instruction
説明
Conversational agents such as those hosted by the 'smart speakers' have become popular over the last few years. Although users can accomplish tasks as if they were asking a person, users still have problems in utilizing conversational agents effectively. To address this problem, some proposals explain how to input agent requests by using visual information such as instruction manuals and displays. However, such instructions create problems such as occupying the hands and eyes. The purpose of this study is to effectively enhance request entry by issuing instructions for use in an easy-to-understand manner without using visual information. Our proposal uses a pair of conversational agents, one is called the main agent, and the other is called the sub-agent, that have different voice types. Experiments show that agent pairing yields easier to understand instructions than using just the main agent. Furthermore, experiments also show that use instructions are easier to understand if the sub-agent reads aloud specific examples of use.
収録刊行物
-
- Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems
-
Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems 1-6, 2019-05-02
ACM