Christian Kroos and Yin Cao together with Phil Coleman from the Institute of Sound Recording have been awarded one of University of Surrey’s Urban Living Award grants.

The project aims to determine the degree of reliability with which an open window can be recognised by humans and machines based only on acoustic cues and to investigate whether the findings for humans and machines can inform each other. To that end a small data base will be recorded, a perception experiment conducted and a deep learning-based recognition system developed. In addition, a machine learning challenge similar to the one ran successfully last year in this project will be organised.

Besides enabling applications in the area of surveillance and home monitoring, the project could become a stepping stone for further projects such as a city-wide app-based indoor sound monitoring project or a project developing personal silent zones (e.g., around the head of a sleeping person) extending existing research at CVSSP and IoSR.

  ⤧  Next post DCASE 2018 Workshop - Looking back ⤧  Previous post Kong et al. (2019a)