SS-SLA-4.5

MULTICHANNEL AUDIO SOURCE SEPARATION WITH INDEPENDENT DEEPLY LEARNED MATRIX ANALYSIS USING PRODUCT OF SOURCE MODELS

Takuya Hasumi, Tomohiko Nakamura, Norihiro Takamune, Hiroshi Saruwatari, The University of Tokyo, Japan; Daichi Kitamura, National Institute of Technology, Kagawa College, Japan; Yu Takahashi, Kazunobu Kondo, Yamaha Corporation, Japan
Session:
Advanced Signal Processing and Machine Learning for Audio and Speech Applications
Track:
Speech, Language, and Audio (SLA)
Session Time:
Fri, 17 Dec, 10:20 - 12:20 Japan Standard Time (UTC +9)
Fri, 17 Dec, 01:20 - 03:20 Coordinated Universal Time
Thu, 16 Dec, 20:20 - 22:20 Eastern Standard Time (UTC -5)
Thu, 16 Dec, 17:20 - 19:20 Pacific Standard Time (UTC -8)
Session Co-Chairs:
Shoji Makino, Waseda University and Hiroshi Saruwatari, The University of Tokyo and Koichi Shinoda, Tokyo Institute of Technology
Presentation
Not logged in.
Discussion
Not logged in.
Resources
Not logged in.
Session SS-SLA-4
FR1.LS-B.1: Development of a Synthetic Database for Compact Neural Network Classification of Acoustic Scenes in Dementia Care Environments
Abigail Copiaco, Nidhal Abdulaziz, University of Wollongong in Dubai, United Arab Emirates; Christian Ritz, University of Wollongong, Australia; Stefano Fasciani, University of Oslo, Norway
FR1.LS-B.2: Reducing algorithmic delay using Low-overlap window for online Wave-U-Net
Sotaro Nakaoka, Takeshi Yamada, Tsukuba University, Japan; Li Li, NTT Communication Science Laboratories, Japan; Shoji Makino, Waseda University, Japan
FR1.LS-B.3: FRAMEWISE FINITE IMPULSE RESPONSE FILTERING BASED ON TIME-FREQUENCY MASK FOR LOW-LATENCY SPEECH ENHANCEMENT
Chiho Haruta, Nobutaka Ono, Yuma Kinoshita, Tokyo Metropolitan University, Japan
FR1.LS-B.4: CONSTRAINED MAXIMUM DIRECTIVITY BEAMFORMERS BASED ON UNIFORM LINEAR ACOUSTIC VECTOR SENSOR ARRAYS
Xueqin Luo, Jilu Jin, Jingdong Chen, Wen Zhang, Northwestern Polytechnical University, China; Gongping Huang, Israel Cohen, Israel Institute of Technology, Israel; Jacob Benesty, University of Quebec, Canada
FR1.LS-B.5: MULTICHANNEL AUDIO SOURCE SEPARATION WITH INDEPENDENT DEEPLY LEARNED MATRIX ANALYSIS USING PRODUCT OF SOURCE MODELS
Takuya Hasumi, Tomohiko Nakamura, Norihiro Takamune, Hiroshi Saruwatari, The University of Tokyo, Japan; Daichi Kitamura, National Institute of Technology, Kagawa College, Japan; Yu Takahashi, Kazunobu Kondo, Yamaha Corporation, Japan