Inclusivity in Communication: Exploring Social Robots for Cultural Integration and Stuttering Therapy

Abstract

Communication is fundamental for humans to exchange ideas, interact, and collaborate. These factors make it important to explore human-robot interactions through the lens of effective communication. This work explores inclusivity from two key dimensions: cultural integration and speech impairments.

Human-robot interaction (HRI) research has increasingly recognised the influence of culture on human-robot interactions, highlighting both the opportunities it brings and the need for careful consideration due to the evolving nature of cultural dynamics. In an online study with 103 participants, we explored how preferences for cross-cultural greetings performed by a humanoid robot change based on a restaurant theme. We examined factors influencing these preferences by analyzing two groups who experienced different ethnic greetings. We studied how ethnicity, percentage of life lived in Western countries, personality characteristics, and implementation of cultural aspects influenced the likability of a robot’s greeting gestures.

In the context of speech impairments, the use of social robots in clinical settings for stuttering therapy of preschoolers by Speech-Language Pathologists (SLPs) remains a relatively unexplored area. To help HRI researchers explore this field, we identified potential applications of social robots for stuttering therapy. We discussed key considerations for these interactions by conducting interviews and shadowing sessions with two SLPs. The results suggest practical applications and interaction and communication design considerations for integrating social robots into speech therapy sessions for preschoolers who stutter. Key applications for social robots include engaging children through various games and providing a consistent speech model, among others. Essential properties of interaction design for sessions include being interactive, structured, and supervised by SLPs. Additionally, the robot’s communication could involve slow, relaxed, personalised speech and other techniques to support children’s speech goals.

Following recommendations from a handbook on stuttering and SLPs, we explored an application area focused on slow and relaxed speech that is to be used by parents and caregivers of children who stutter. As a first step and proof-of-concept, I carried out an HRI study with university students. I designed two sessions with four and five tasks, collecting natural speech samples from 63 participants across different verbal tasks. Throughout the interactions, the voice system (two conditions: the robot speaking or the same voice coming from a laptop) maintained a slow speech. I report our results on which kind of tasks had more impact on participants’ speech and the difference in participants’ speech metrics in the two sessions. People with different English language-related self-reported factors also showed varied patterns in adapting to the Voice System’s slow speech, showcasing the need for customised implementations in the future.

This work contributes to HRI by highlighting the challenges of developing culturally adaptive robots and providing design considerations for using social robots in stuttering therapy for preschoolers. By studying speech entrainment through a Voice System exhibiting slow speech, we explored how different verbal tasks influence entrainment levels, with potential future applications in training parents and caregivers of children who stutter.