Exploring the Impact of Error Feedback Methods on User Experience in Voice Interaction

Abstract

Voice interaction has played an important role in various scenarios such as smart homes, cars, and healthcare due to its ease of use, efficiency, and convenience. However, errors in voice interaction can greatly affect the user experience. This paper aims to explore the impact of different error feedback methods on user experience, with the goal of improving the user experience of voice interaction. The study utilizes a combination of subjective and objective approaches by creating an experimental platform to collect facial expression data and emotional valence evaluation data from participants. By analyzing the data, user preferences for different feedback methods can be determined. The findings suggest that in directive task scenarios, users prefer feedback that directly explains the error. In broadcast task scenarios, users prefer feedback that explains the error and provides a commitment to resolve it. In conversational task scenarios, users prefer intelligent voice assistants to take the lead in the conversation, guide its direction, and provide specific event suggestions. This research contributes to a better understanding of the impact of error feedback methods on user experience and provides guidance and reference for the design of future error feedback methods in voice interaction.

Keywords: voice user interaction, error feedback, emotion recognition, emotion value

DOI: 10.54941/ahfe1005798