Pervasive audio playback in cyber-physical environments

Although devices in cyber-physical environments — settings that are rich of connected sensors and actuators — are getting more powerful in terms of speed and connection capabilities, they still barely communicate with each other. By thinking of their functionality as a service rather than treating each device as a separate unit, new context-aware applications in several domains could be realized. This paper presents a framework for audio playback in cyber-physical environments named ‘Audio as a Service’. It features a method for loudspeaker selection and automatic adjustment of the audio output volume. This allows pervasive audio presentation in conjunction with user position data, while playing synchronously to an arbitrary number of connected loudspeakers. It further implements device discovery and a flexible interface to process and distribute auditive content. We present the service architecture and loudspeaker selection method, as well as results of a study showing a news reader application based on the system.