Seminar by Raphaël LEIBA

Large array of microphones for the automatic estimation of the noise annoyance induced by road traffic in urban environments

Road traffic noise represent a major part in noise annoyance in city centres. In order to increase city dwellers wellness, it can be interesting to diagnose the noise annoyance induced by each road vehicle to feed into city planners and decision makers workflow. In this work, microphone arrays and beamforming techniques are used in order to extract the audio signal of each vehicles in the road traffic. This is done with the help of camera based tracking systems to localise each vehicles. Psychoacoustic indices and the knowledge of the vehicle type and speed, obtained via machine learning, are then used to estimated each vehicle-induced noise annoyance.
Proof of concept results and a validation of the noise annoyance model via a psychoacoustic experiment will be presented.

Keywords : Machine learning, beamforming, vehicle detection, moving source signal filtering, psychoacoustic experiments, noise annoyance.