Researchers take inspiration from KGB to develop context aware devices

The group’s work – which is being presented at the Association for Computing Machinery's User Interface Software and Technology Symposium in Berlin – centres on two approaches: one which makes innovative use of the microphones present in most devices and the other employing a modern-day version of eavesdropping technology used by the KGB in the 1950s.

The first of these, a sound-based activity recognition system called Ubicoustics, uses sound-effect libraries to train existing microphones to enable devices to recognise sounds associated with places, such as bedrooms, kitchens, workshops, entrances and offices.

"The main idea here is to leverage the professional sound-effect libraries typically used in the entertainment industry," explained Gierad Laput, a PhD student in the group. "They are clean, properly labelled, well-segmented and diverse. Plus, we can transform and project them into hundreds of different variations, creating volumes of data perfect for training deep-learning models."