Research Area

  • Title:
    Environmental Audio Analysis in real-world scenarios

  • Keywords:

    Deep Neural Networks, Acoustic Scene Classification, Novelty Detection

  • Description:

    Different research fields rely on Digital Signal Processing and Machine Learning techniques in order to face complex modelling, prediction, and recognition tasks. One of these is represented by Digital Audio, which finds application in entertainment, security, forensics and health to name but a few. The typical methodology adopted in these solutions consists in extracting and manipulating useful information from the audio streams captured my means of multiple microphones distributed in the environment to be monitored.
    The goal of this research is to employ Deep Neural Networks for acoustic monitoring and analysis of acquired audio signals in a living environment, addressing different tasks such as audio event detection, novelty detection, audio source localization. The accomplishment of these tasks and the gathering of related information about the status and characteristics of the environmental soundscape is a fundamental to pilot the execution of specific automated services.

  • Laboratory:

    Computational Audio Processing Lab

  • Contact Person:

    Stefano Squartini

  • Collaborations:

    MAC Srl, ASK Industries, FairConnect, Tampere University of Technology, University of Passau, University of Sheffield

  • Projects:
    • SINC (Progetto POR 2018-2020), MOHMI – MIRACLE (Piattaforma Domotica 2020)
    • Private funding of PhD and PostDoc Research Fellowships
    • Challenge
    • Vitality