Accelerating the Development of Machine Voice User Interfaces with Immersive Environments

This paper proposes a conceptual framework leveraging VR technologies to develop large language model (LLM)-based Voice User Interfaces (VUIs) for industrial applications, addressing key adoption challenges. By integrating virtual twins and immersive environments, the framework supports essential development steps, including data collection, LLM training, usability testing, and iterative design validation. The study shows how VR enables realistic simulations, faster development, and reduces the dependency on physical prototypes in the VUI development, offering improvements in efficiency and user-centered design. Initial findings of the proof-of-concept study highlight the framework’s potential to address industrial requirements.