Non-Humanoid Robotics Speech Synthesizer: Design and User Perception

Abstract

This paper introduces a Non-Humanoid Robotics Speech Synthesizer (NRSS) designed for identifying and greeting individuals through verbal responses. The system integrates a text-tospeech module, ultrasonic sensor, control circuitry, and a microcontroller, all programmed using Arduino. This setup enables efficient person identification, greetings, and information relay. Building on previous studies about robot speech perception, this study explores user reactions to the NRSS during initial encounters. The premise is that non-humanoid robots with verbal capabilities can be positively perceived by users. To test this, 54 participants of various ages and robotic experiences evaluated the NRSS. Using one-way ANOVA, the study found that the NRSS was positively perceived in terms of competence and comfort across demographics, although perceptions of warmth varied. The study concludes that the NRSS excels in environments where perceived competence and user comfort are paramount, such as public greeting or guidance roles, due to its low cost and ease of construction, though its utility can extend to various other domains.