It´s not only what you say, but also how you say it! How expressed Emotions predict User’s Satisfaction with Voice Assistants in different Usage Contexts
(A2025-124410)

AUTHORS

John Vara Prasad Ravi, IQS School of Management - Universitat Ramon Llull; Jan-Hinrich Meyer, Universitat Ramon Llull, IQS School of Management; Ramon Palau-Saumell, IQS School of Management Barcelona; Divya Seernani, iMotions A/S

ABSTRACT

Voice assistants have reshaped customer service by offering new interaction channels. This study explores how user emotions during interactions with multimodal and voice-only devices across different contexts affect satisfaction. Using a novel approach throughout three experimental studies that captures the user´s experienced emotions via voice tone and speech content analysis, we show that both device type and context are crucial in shaping user emotions and satisfaction. The first and second studies investigate the pleasantness and complexity of the task. The third study explores the role of device anthropomorphism in eliciting consumer emotions and satisfaction. We used a novel voice algorithm and text-mining procedures to extract objective multidimensional measures of user-expressed emotions during the interaction. The emotions expressed via voice tone or speech content can explain differences in user satisfaction and could therefore be useful in enhancing the user experience.