A Conversational Agent for Cultural and Environmental Education: Enhancing User Experience and Immersion in Augmented Reality
摘要
Augmented Reality (AR) and Artificial Intelligence (AI) have transformed education by offering innovative tools that enhance interactivity, engagement, and tailor experiences to users’ preferences. This paper explores the integration of a conversational agent into an AR application designed to guide and educate users on the cultural history and environmental conditions of eastern Crete. The study focuses on integrating a conversational agent into an AR application, utilizing speech-to-text conversion, large language model processing, and text-to-speech capabilities to facilitate interactive experiences. In addition to Generative AI for generating responses, pre-defined scripts are included for quicker information delivery in certain events. A pilot study was conducted to evaluate user experience, cognitive load, and social agency, revealing high satisfaction and engagement with the application. The application was tested in real-world environments, using surveys assessing user engagement, cognitive load, and social agency, including the NASA-TLX, User Engagement Scale, and Social Presence questionnaires. The results showed high user satisfaction, with positive feedback across all evaluation aspects. Participants found the conversational agent helpful and effective, though suggestions for improvement included adding facial expressions to the avatar and refining the user interface.