An architecture for real-time interactive music performances

We present a system architecture that delivers real-time Extended Reality (XR) music performances. Live shows are captured using multichannel audio and depth-enhanced video, then processed and streamed through a distributed media platform that supports low-latency audio rendering and dynamic scene composition. This approach balances performance with scalability and is suitable for deployment on standard infrastructure. Our findings demonstrate that the system can bridge physical distances, support audience feedback, and provide an engaging, interactive concert experience that is scalable across diverse environments.