Fri AM1.L5.3
Constant Separating Vector-based Blind Source Extraction and Dereverberation for a Moving Speaker
Tetsuya Ueda, Shoji Makino, Waseda University, Japan
Session:
Fri AM1.L5: Advanced Signal Processing and Machine Learning for Acoustic Scene Analysis and Signal Enhancement Lecture
Track:
Special Sessions
Location:
Press room
Presentation Time:
Fri, 8 Sep, 11:10 - 11:30 Finland Time (UTC +3)
Session Co-Chairs:
Shoji Makino, Waseda University and Nobutaka Ono, Tokyo Metropolitan University
Presentation
Discussion
Resources
No resources available.
Session Fri AM1.L5
Fri AM1.L5.1: LOCALIZING SPATIAL INFORMATION IN NEURAL SPATIOSPECTRAL FILTERS
Annika Briegleb, Thomas Haubner, Vasileios Belagiannis, Walter Kellermann, Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany
Fri AM1.L5.2: NoisyILRMA: Diffuse-Noise-Aware Independent Low-Rank Matrix Analysis for Fast Blind Source Extraction
Koki Nishida, Norihiro Takamune, The University of Tokyo, Japan; Rintaro Ikeshita, NTT Corporation, Japan; Daichi Kitamura, National Institute of Technology, Japan; Hiroshi Saruwatari, The University of Tokyo, Japan; Tomohiro Nakatani, NTT Corporation, Japan
Fri AM1.L5.3: Constant Separating Vector-based Blind Source Extraction and Dereverberation for a Moving Speaker
Tetsuya Ueda, Shoji Makino, Waseda University, Japan
Fri AM1.L5.4: Acoustic Traffic Monitoring Based on Deep Neural Network Trained by Stereo-Recorded Sound and Sensor Data
Tomohiro Takahashi, Yuma Kinoshita, Yukoh Wakabayashi, Nobutaka Ono, Tokyo Metropolitan University, Japan; Jun Honda, Seishi Fukuma, Aoi Kitamori, Hiroshi Nakagawa, NEXCO-EAST ENGINEERING, Japan
Fri AM1.L5.5: Independent Vector Extraction Constrained on Manifold of Half-Length Filters
Zbynek Koldovsky, Jaroslav Cmejla, Technical University of Liberec, Czech Republic; Tulay Adali, University of Maryland, United States; Stephen O'Regan, Naval Surface Warfare Center Carderock Division, United States