University of Surrey Scientists Celebrate Success of Top-ranking ‘Listening’ Technology

Published on: 17 Sep, 2018
Updated on: 19 Sep, 2018

By Will Parsons

Researchers at the University of Surrey have created a system that can recognise everyday sounds, such as a dog barking or a guitar strumming. The technology was developed for the Detection and Classification of Acoustic Scenes and Events (DCASE) 2018 challenge, in which it was ranked third out of 558 systems created worldwide.

As part of the SCASE 2018 challenge, the Kaggle Freesound Audio Tagging Challenge tasks competitors with creating a machine that can identify and understand everyday noises.

Team members from the university’s Centre for Vision, Speech and Signal Processing (CVSSP) outperformed major international corporations and global university groups to reach the top of the leaderboard.

Dr Wenwu Wang, a CVSSP team member and audio researcher at the university, said: “This is an outstanding achievement following our success on the DCASE 2017 Challenge, for which we also topped the ranking list. This further confirms our world-leading research in sound recognition.”

CVSSP’s audio tagging system uses artificial intelligence (AI) to emulate human hearing, enabling the machine to have better situational awareness of sounds and improving its decision making. The university claims that such technology could have many applications ranging from assisted living to defence.

Professor Adrian Hilton, Head of CVSSP, added: “This is an excellent achievement for the team… Our challenge over the next decade is to bring these technologies together to enable future intelligent systems for healthcare, robotics, automotive and entertainment sectors.”