Speaker: Archontis Politis
Data and place: Sep 29, 2022, at 10:30 – Hybrid
Abstract: Archontis Politis is a researcher on spatial audio technologies currently at Tampere University, Finland, and in close collaboration with Aalto University, Finland. This presentation summarizes work that the researcher has been involved in those two universities, mainly around two areas. The first is on parametric spatial audio processing, meaning techniques that process multi-microphone recordings, extract spatial parameters, and use those parameters to provide an informed spatial rendering to flexible playback setups or to perform spatial enhancement or modifications of the recorded content. The second area is on acoustic scene analysis using machine learning techniques, and more specifically on simultaneous localization, detection, and classification of sound events. This research problem has attracted significant interest as a recent task in the DCASE Challenge, with the presenter being its main organizer. An overview of the evolution of the task and open questions related to this problem is presented.
Speaker’s bio: Archontis Politis is an assistant professor at Tampere University, Finland. He obtained his M.Sc. degree in Sound & Vibration studies from the Institute of Sound and Vibration Research (ISVR), University of Southampton, UK, in 2008. In 2015 he was a visiting researcher at the University of Maryland Institute for Advanced Computer Studies, MA, USA, and in the same year he completed a research internship at Microsoft Research, Redmond, WA, USA. In 2016 he obtained a Doctor of Science degree on spatial audio processing from Aalto University, Finland. He has served as editor of a book on Parametric Spatial Audio Processing, organizer in the DCASE scientific challenge, and has chaired various special sessions in international conferences. His research interests include spatial audio technologies, virtual acoustics, array signal processing, and acoustic scene analysis.