Human Perception of Emotional Responses to Changes in Auditory Attributes of Humanoid Agents

Human-robot interaction has emerged as an increasingly prominent discourse within the domain of robotic technologies. In this context, the interaction of both visual and verbal cues assumes an essential role in shaping user experiences. The research problem of this study revolves around investigating the potential impact of auditory attribute alterations in humanoid agents, namely robots and avatars, on users’ emotional responses. The study recruited a participant cohort comprising 14 individuals, aged 18 to 35, to engage in an experimental process of observing avatar videos with distinctive auditory attributes. These attributes encompassed two voice pitches, specifically alto voice and bass voice, as well as two speech styles denoted as frozen style and casual style. Through data collection, a repository of 13,600 data points was amassed from the participants, and subsequently subjected to rigorous analysis via the ANOVA methodology. The empirical findings demonstrate that users reveal sensitive, emotional responsiveness when faced with avatar videos characterized by varying auditory attributes. This pilot study establishes a foundational framework poised to guide future research undertakings aimed at inspiring user experiences through the deliberate manipulation of auditory attributes inherent in humanoid robots and avatars.