This project will investigate a device for recording and analysing environmental sounds in an urban setting. The goal is to detect particular vocal species and to report on them through an engaging interface.
Research will involve designing software for the device to process sound in real time to detect species of interest. This will using machine learning techniques and will run on a Raspberry Pi based device. Research will also encompass how to make the output suitable for citizen scientists and engagement.
The outcome will be a prototype software systems to record and analyse sound. It may also encompass the refinement of the Raspberry Pi based sensor hardware and networking.
Skills and experience
Students need experience in Raspberry Pis, plus some software and hardware experience and a willingness and interest to learn new things including investigating a machine learning framework for analysing sound.
Contact the supervisor for more information.