Spacify

Traditionally, designing interiors, buildings, or urban spaces has required advanced technical and visuospatial skills, limiting participation to trained professionals. In this independent study, we developed Spacify, a prototypical framework designed to make spatial design accessible to non-experts through natural language. Leveraging advances in large language models (LLMs) and extended reality (XR), the framework integrates five core components—External Data, User Input, Spatial Interface, Large Language Model, and Current Spatial Design—to enable users to understand, articulate, and visualize 3D spaces through plain-language interaction. Demonstrated via an XR smartphone application, the system supports question-and-answer exchanges about 3D environments, (re)generation of designs from natural language prompts, and visualization of spatial concepts described in text. Rapid user testing showed that Spacify lowers barriers to participation, expands access to 3D spatial design, and introduces a language-first approach to built environment creation.