Acoustic Signal Processing today – Merging Physical Models and Machine Learning

Traditional methods for longstanding acoustic signal processing challenges, such as speech enhancement, echo control, or source localization, relied for a long time on more or less sophisticated physical modeling for the involved signals and systems, typically complemented by – often empirical - methods to efficiently estimate model parameters. Fueled by tremendous progress in signal processing theory and hardware, and reinforced by commercially successful products, increasingly complex tasks can be tackled and advanced machine learning algorithms become viable options for complementing or even replacing traditional model-based methods. In this talk, we will argue for efficient combinations of model-based and machine learning algorithms that minimize resource consumption and maximize interpretability, and present examples in the areas of echo cancellation, noise suppression, signal extraction and acoustic sensor networks.

Short Bio: Walter Kellermann is a professor for communications at the University of Erlangen-Nuremberg, Germany, since 1999. He received the Dipl.-Ing. (univ.) degree in Electrical Engineering from the University of Erlangen-Nuremberg, in 1983, and the Dr.-Ing. degree from the Technical University Darmstadt, Germany, in 1988. From 1989 to 1990, he was a postdoctoral Member of Technical Staff at AT&T Bell Laboratories, Murray Hill, NJ. In 1990, he joined Philips Kommunikations Industrie, Nuremberg, Germany, to work on hands-free communication in cars. From 1993 to 1999, he was a Professor at the Fachhochschule Regensburg, where he also became Director of the Institute of Applied Research in 1997. In 1999, he cofounded DSP Solutions, a consulting firm in digital signal processing, and he joined the University Erlangen-Nuremberg as a Professor and Head of the Audio Research Laboratory. He authored or coauthored 20+ book chapters, 300+ refereed papers in journals and conference proceedings, as well as 70+ patents, and is a co-recipient of ten best paper awards. His current research interests include speech signal processing, array signal processing, adaptive filtering, and its applications to acoustic human–machine interfaces. Dr. Kellermann served as an Associate Editor and  Guest Editor to various journals, including the IEEE Transactions on Speech and Audio Processing from 2000 to 2004, the  IEEE Signal Processing Magazine in 2015, and presently serves as Associate Editor to the EURASIP Journal on Applied Signal Processing. He was the General Chair of eight mostly IEEE-sponsored workshops and conferences. He served as a Distinguished Lecturer of the IEEE Signal Processing Society (SPS) from 2007 to 2008. He was the Chair of the IEEE SPS Technical Committee for Audio and Acoustic Signal Processing from 2008 to 2010, a Member of the IEEE James L. Flanagan Award Committee from 2011 to 2014, a Member of the SPS Board of Governors (2013-2015), Vice President Technical Directions of the IEEE Signal Processing Society (2016-2018) and is currently a member of the SPS Nominations Appointments Committee (2019-2022). He was awarded the Julius von Haast Fellowship by the Royal Society of New  Zealand in 2012 and the Group Technical Achievement Award of the European Association for Signal Processing (EURASIP) in 2015. In 2016, he was a Visiting Fellow at Australian National University, Canberra, Australia. He is an IEEE Fellow and was elevated to EURASIP Fellow in 2021.