Arkay is an AI-powered tour assistant designed to transform how we explore architectural and cultural landmarks. Unlike conventional guidebooks, expensive personal tours, or static audio guides, Arkay offers a dynamic, personalized, and interactive discovery experience—accessible to anyone with a smartphone or smart glasses.

What is Arkay?

Arkay is a real-time, vision-language model-based assistant that helps users understand the buildings and sites they encounter. By simply pointing a camera at a structure, users can receive AI-generated information about its architectural style, historical background, structural features, and cultural significance. The system doesn't rely on preloaded data but analyzes visual input on the spot, delivering context-aware interpretations and inviting the user into a two-way dialogue.

Whether you are standing in front of a medieval cathedral, a modernist pavilion, or an ancient ruin, Arkay provides:

• Descriptions of architectural form, style, construction strategy

• Contextual analysis of urban placement and cultural background

• Real-time interaction through voice or text

Why It Matters

Every year, more than 1.3 billion tourists travel internationally. Many of them seek deeper knowledge about the architectural and cultural spaces they visit but face limitations in accessibility, cost, and depth of available information.

Traditional guides are costly and inflexible, audio systems disrupt the experience, and generic AI lacks architectural depth. Every year, more than 1.3 billion tourists travel internationally. Many of them seek deeper knowledge about the architectural and cultural spaces they visit but face limitations in accessibility, cost, and depth of available information. Traditional guides offer limited flexibility. AR/VR platforms, though promising, are costly and not widely available. General AI tools lack the architectural fluency required for in-depth interpretation.

Arkay fills this gap by offering:

• Comprehensive and verified information

• Reduced reliance on printed materials and resource-heavy guide systems

• Low-cost, scalable access to architectural knowledge

• Real-time commentary tailored to the user’s location and interests

• Portable usage via mobile app or smart glasses

• Culturally embedded, expert-informed content delivery
  
  

How It Works

At its core, Arkay operates as a lightweight, cloud-based agent system that listens to user queries, processes them through a trained AI model, and responds with both text and synthesized voice. The system's components include:

• A Vision-Language Model (VLM) that interprets architectural imagery

• A Text-to-Speech (TTS) engine for audio feedback

• GPS and camera integration to provide location-based contextual understanding

Users can simply ask:

"Hey Arkay, tell me about this building."

And receive:

"This is a Brutalist government building from the 1970s, notable for its raw concrete façade, deep-set windows, and emphasis on structural expression."

Deployment & Impact

Arkay is currently in the prototype phase, with MVP and agent systems under development for deployment on mobile and smart eyewear platforms. Pilot tests are planned for culturally rich locations in Izmir, including archaeological sites and museums.

The project is supported by the UMAI Lab and received application backing through the TUBITAK BIGG initiative. Arkay aligns with several UN Sustainable Development Goals (SDGs):

• SDG 4: Quality Education through informal cultural learning

• SDG 8: Decent Work and Economic Growth in cultural tourism

• SDG 9: Industry, Innovation, and Infrastructure via AI-driven urban experience

Future Vision

Arkay is more than a tour guide. It is a speculative interface between people, place, and knowledge—designed to evolve with the cities it inhabits. With growing smart tourism demand, Arkay paves the way for AI-augmented cultural encounters, making architectural knowledge both personal and public.

Explore. Ask. Learn. Wherever you are, Arkay is by your side.