Empathetic conversations in the car

UPDATED—February 13, 2019.Mental health is a world-wide problem impacting 400million people and costing 1 trillion annually [1], but currently, providers of mental health therapies and services are not adequately meeting the needs of the patient population. In this work, we discuss user interactions with voice chat agents and limitations of current system designs in facilitating a coherent user experience with these agents. This is especially crucial, if voice agents need to convey the conversational flow required in applications such therapeutic chat bots. Additionally, the use of back-channels, or filler phrases or sounds such as uh-huh or ah, could provide users a more intuitive experience with therapeutic chat bots. Through a series of user experience tests, we present a preliminary data set of human-intuited back-channel insertions and turn-taking detection for future use in developing user interactions with conversational agents. CHI’19, May 2019, Glasglow, UK © 2019 Association for Computing Machinery. This is the author’s version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in Proceedings of ACM CHI Conference (CHI’19), https://doi.org/https://doi.org/10.1145/3290607. Empathetic conversations in the car CHI’19, May 2019, Glasglow, UK ACM Reference Format: Hiroshi Mendoza, Julia Alison, Shannon Wu, James Landay, and Pablo E. Paredes. 1997. Empathetic conversations in the car. In Proceedings of ACMCHI Conference (CHI’19).ACM, New York, NY, USA, 6 pages. https://doi.org/https: //doi.org/10.1145/3290607