The Development of Architectural Approaches for Creating Interactive Scenarios for VR and AR Use

Background: As immersive virtual reality (VR) and augmented reality (AR) environments evolve, their architectural frameworks must adapt to support dynamic interactivity and user agency.

System Overview: This manuscript introduces the VIODET platform, which advances the design of interactive scenarios by offering two complementary architectural approaches unified under a single cross-platform Unity client. The platform deploys seamlessly across the Meta Quest 3, mobile devices in Mixed Reality environments, and Windows PCs.

Architectural Approaches:

?    AI-Powered Dynamic Interaction: The first approach leverages cloud-based artificial intelligence, including Speech-to-Text, Large Language Models, and Text-to-Speech, to facilitate natural, free-form dialogues and adaptive narrative progression. It utilizes a dual-stream animation system that blends cloud-driven lip synchronization with locally managed body movements, effectively mitigating network latency artifacts.

?    Non-AI Deterministic Authoring: The second approach relies on a backend graphical scenario design tool to create predetermined multiple-choice dialogue graphs and event mappings. This deterministic framework guarantees complete outcome predictability and testability, making it highly suitable for compliance-driven applications, medical simulations, and high-stakes training.

Conclusion: By decoupling content creation from the runtime environment via a modular client-server model, VIODET successfully reconciles the demand for emergent, exploratory user experiences with the necessity for authorial control and production efficiency. This dual-architecture strategy serves as a resilient and accessible framework for future extended reality (XR) development across educational, entertainment, and industrial domains.