Thursday, Sept. 8

9:00 – 10:00 Keynote 3
Plenary Hall Chair: Patrick Naylor

Spatial Acquisition, Digital Archiving, and Interactive Auralization
Toon van Waterschoot
KU Leuven
10:00 – 10:30 Coffee Break
Poster Area  
10:30 – 12:30  Poster Session E
Poster Area
Chair: Christiane Antweiler
E-01 AID: Open-Source Anechoic Interferer Dataset
Philipp Götz1, Cagdas Tuna2, Andreas Walther2, and Emanuël A. P. Habets1
1International Audio Laboratories Erlangen, Germany
2Fraunhofer Institute for Integrated Circuits Erlangen, Germany
E-02 Acoustic Echo Suppression using a Learning-based Multi-Frame Minimum Variance Distortionless Response (MFMVDR) Filter
Yuefeng Tsai, Yicheng Hsu, and Mingsian Bai
National Tsing Hua University, Taiwan
E-03 Source Separation for Sound Event Detection in Domestic Environments using Jointly Trained Models
Diego de Benito-Gorrón1, Kateřina Žmolíková2, and Doroteo T. Toledano1
1Universidad Autónoma de Madrid, Spain
2Brno University of Technology, Czechia
E-04 Independent Vector Analysis Assisted Adaptive Beamforming for Speech Source Separation on an Acoustic Vector Sensor
Yichen Yang, Xianrui Wang, Wen Zhang, and Jingdong Chen
1Northwestern Polytechnical University, China
E-05 3D Single Source Localization Based on Euclidean Distance Matrices
Klaus Brümann and Simon Doclo
University of Oldenburg, Germany
E-06 Phase Error Analysis for First-Order Linear Differential Microphone Arrays
Longfei Yan1, Weilong Huang2, W. Bastiaan Kleijn1, and Thushara D. Abhayapala3
1Victoria University of Wellington, New Zealand
2Alibaba Group, China
3Australian National University, Australia
E-07 Training Strategies for Own Voice Reconstruction in Hearing Protection Devices using an In-ear Microphone
Mattes Ohlenbusch1, Christian Rollwage1, and Simon Doclo2
1Fraunhofer IDMT, Germany
2University of Oldenburg, Germany
E-08 Two-Stage Speech Enhancement Using Gated Convolutions
Lars Thieling and Peter Jax
RWTH Aachen University, Germany
E-09 Accelerated Unsupervised Clustering in Acoustic Sensor Networks using Federated Learning and a Variational Autoencoder
Luca Becker, Alexandru Nelus, Rene Glitza, and Rainer Martin
Ruhr-Universität Bochum, Germany
E-10 Positional Tracking of a Moving Microphone in Reverberant Scenes by Applying Perfect Sequences to Distributed Loudspeakers
Fabrice Katzberg, Marco Maass, René Pallenberg, and Alfred Mertins
University of Lübeck, Germany
E-11 Echo Cancellation and Noise Suppression by Training a Dual-Stream Recurrent Network with a Mixture of Training Targets
Fatemeh Alishahi, Yin Cao, Youngkoen Kim, and Asif Mohammad
Qualcomm Technologies, USA
E-12 Task Splitting for DNN-based Acoustic Echo and Noise Removal
Sebastian Braun and Maria Luis Valero
Microsoft, USA/Germany
E-13 Fixed Beamformer Design Using Polynomial Eigenvalue Decomposition
Vincent W. Neo, Emilie d’Olne, Alastair H. Moore, and Patrick A. Naylor
Imperial College London, UK
E-14 Realistic Sources, Receivers and Walls Improve the Generalisability of Virtually-Supervised Blind Acoustic Parameter Estimators
Prerak Srivastava, Antoine Deleforge, and Emmanuel Vincent
INRIA Nancy, France
10:30 – 12:30 Demonstrations B
  Chair: Henning Puder
Hearing Aids Connected to the World of Sensors and Apps
Henning Puder and Stefan Petrausch
WS Audiology, Germany
Room K8
Networked Robots for Remote Dynamic Acoustic Experiments
Ethaniel Moore, Austin Lu, George Zhai, Manan Mittal, Kanad Sarkar, Ryan M. Corey, Paris Smaragdis, and Andrew Singer
University of Illinois at Urbana-Champaign, USA
Room K3
Mobile, Multi-Sensor, Real-Time Signal Processing Setup for Synchronous Recordings in Real-Life Situations
Kamil Adiloğlu1, Lisa Straetmans2, Micha Lundbeck1, Paul Maanen2, Mats Exter1, Stefan Debener2
1Hörzentrum Oldenburg, Germany
2University of Oldenburg, Germany
12:30 – 14:00 Lunch Break
Lunch Room
14:00 – 16:00  Poster Session F
Poster Area Chair: Sharon Gannot
F-01 CPTNN: Cross-Parallel Transformer Neural Network for Time-Domain Speech Enhancement
Kai Wang, Bengbeng He, and Wei-Ping Zhu
Concordia University, Canada
F-02 Bandwidth-Scalable Fully Mask-Based Deep FCRN Acoustic Echo Cancellation and Postfiltering
Ernst Seidel1, Rasmus Kongsgaard Olsson2, Karim Haddad2, Zhengyang Li1, Pejman Mowlaee2, and Tim Fingscheidt1
1Technische Universität Braunschweig, Germany
2GN Audio A/S, Denmark
F-03 A Bilinear Framework for Adaptive Speech Dereverberation Combining Beamforming and Linear Prediction
Wenxing Yang1, Gongping Huang2,4, Andreas Brendel4, Jingdong Chen1, Jacob Benesty3, Walter Kellermann4, and Israel Cohen2
1Northwestern Polytechnical University, China
2Technion – Israel Institute of Technology, Israel
3University of Quebec, Canada
4Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany
F-04 Array Geometry Optimization for Region-of-Interest Broadband Beamforming
Yuval Konforti, Israel Cohen, and Baruch Berdugo
Technion – Israel Institute of Technology, Israel
F-05 Dual-Compression Neural Network with Optimized Output Weighting for Improved Single-Channel Speech Enhancement
Stefan Thaleiser1, Aleksej Chinaev2, Rainer Martin1, and Gerald Enzner2
1Ruhr-Universität Bochum, Germany
2University of Oldenburg, Germany
F-06 Numerical Investigation of Weight Parameters for Geometrically Constrained Independent Vector Analysis using Vectorwise Coordinate Descent or Iterative Source Steering
Shinya Furunaga1, Kana Goto2, Tetsuya Ueda1, Li Li2, Yamada Takeshi2, and Shoji Makino1
1Waseda University, Japan
2University of Tsukuba, Japan
3NTT Communications and Science Laboratories, Japan
F-07 DeepFilterNet2: Towards Real-Time Speech Enhancement on Embedded Devices for Full-Band Audio
Hendrik Schröter1, Tobias Rosenkranz2, Alberto N. Escalante B.2, and Andreas Maier1
1Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany
2WS Audiology, Germany
F-08 Signal-informed DNN-based DOA Estimation Combining an External Microphone and GCC-PHAT Features
Ulrik Kowalk1, Simon Doclo2, and Jörg Bitzer1
1Jade University of Applied Sciences, Germany
2University of Oldenburg, Germany
F-09 Environmental Sound Classification based on CNN Latent Subspaces
Maha Mahyub1, Lincon S. Souza2, Bojan Batalo1, and Kazuhiro Fukui1
1University of Tsukuba, Japan
2AIST, Japan
F-10 Informed vs. Blind Beamforming in Ad-Hoc Acoustic Sensor Networks for Meeting Transcription
Tobias Gburrek, Jörg Schmalenströer, Jens Heitkämper, and Reinhold Häb-Umbach
Paderborn University, Germany
F-11 Physics-informed Convolutional Neural Network with Bicubic Spline Interpolation for Sound Field Estimation
Kazuhide Shigemi, Shoichi Koyama, Tomohiko Nakamura, and Hiroshi Saruwatari
University of Tokyo, Japan
F-12 An Introduction to the Speech Enhancement for Augmented Reality (SPEAR) Challenge
Pierre Guiraud1, Sina Hafezi1, Patrick A. Naylor1, Alastair H. Moore1, Jacob Donley2, Vladimir Tourbabin2, and Thomas Lunner2
1Imperial College London, UK,
2Reality Labs Research at Meta, USA
F-13 A State-Space Recurrent Neural Network Model for Dynamical Loudspeaker System Identification
Christian Gruber1, Gerald Enzner2, and Rainer Martin3
1voiceINTERconnect, Germany
2University of Oldenburg, Germany
3Ruhr-Universität Bochum, Germany
F-14 MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator
Tobias Cord-Landwehr, Thilo von Neumann, Christoph Böddeker and Reinhold Häb-Umbach
Paderborn University, Germany
14:00 – 16:00 Demonstrations C
  Chair: Henning Puder
Low delay processing for PureSound
Lars Dalskov Mosgaard and David Pelegrin Garcia
WS Audiology, Germany/Denmark
Room K3
Real-time DNN-based Acoustic Echo and Noise Removal
Sebastian Braun
Microsoft, USA
Room K3
Ava: Online Captioning & Speaker Diarization
Alexey Ozerov
Ava, USA/France
16:00-16:30 Award Ceremony and Closing
Plenary Hall
Chair: Walter Kellermann