A social multi-camera interactive live engagement (SMILE) system including video and audio processing arrays performs an audio-video processing method. The method includes receiving multiple video streams from distinct sources and preparing multiple distinct composited video streams from the multiple video streams. Each of the distinct composited video streams configures video images from a different selected one of the multiple video streams in a primary region and video images from remaining ones of the multiple video streams in respective secondary regions adjacent to the primary region. The method include streaming the multiple distinct composited video streams configured for presentation in a user interface to a client device, at different times determined in response to user selection input from the client device, and more detailed aspects, supporting a novel user interface design.