Researchers from Queen Mary University of London’s Centre for Digital Music have teamed up with commercial sound recognition pioneer Audio Analytic, with the support of two new grants worth £385K from the Technology Strategy Board.
Home security and automation systems could benefit from one of the new projects, which will look at improving the recognition of generic environmental sounds.
Audio Analytic’s software recognises sounds automatically, which helps users to detect incidents or events that cannot be identified by other techniques. Through this collaboration between the company and the academic research group, cutting-edge recognition methods and new applications will be explored and added to the system.
“Our algorithms and sensors are already pretty good at recognising music, but being able to detect more generic environmental sounds such as an alarm going off or a loud crash can be very useful if implemented into security systems,” said Professor Mark Plumbley, Director of the Centre for Digital Music at QMUL's School of Electronic Engineering and Computer Science.
The first project will create a small, portable and wireless sensor that can automatically alert the user when detecting certain sounds. Baby monitoring, remote smoke alarm notifications, or even care homes that monitor the safety of the elderly, are commercial applications which could benefit from this kind of ‘smart’ microphone.
The second project will develop new audio analysis and modelling techniques to allow people to index and explore audio data. This could have many applications in professional security or home security as it will allow the indexing of large sound and video databases. It will also be useful in the monitoring of environmental and industrial sounds.
“Audio Analytic is the emerging world leader in automatic environmental sound recognition. Its cutting-edge technology empowers people to use audio as never before. We are delighted to collaborate with Professor Plumbley’s research group, thus teaming up the best of academic expertise with the most advanced commercial know-how to develop unparalleled sound recognition technology,” said Dr Sacha Krstulovic, Lead Research Engineer at Audio Analytic.
Both projects will also be supported by QMUL’s Centre for Intelligent Sensing, which develops advanced sensor technology and applications.
For media information, contact: