DIVERGENCE OPTIMIZATION IN NONNEGATIVE MATRIX FACTORIZATION WITH SPECTROGRAM RESTORATION FOR MULTICHANNEL SIGNAL SEPARATION

被引：0
作者：
Kitamura, Daichi [1 ]
Saruwatari, Hiroshi [1 ]
Nakamura, Satoshi [1 ]
Takahashi, Yu [2 ]
Kondo, Kazunobu [2 ]
Kameoka, Hirokazu [3 ]
机构：
[1] Nara Inst Sci & Technol, Nara 6300101, Japan
[2] Yamaha Corp, Shizuoka, Japan
[3] Univ Tokyo, Tokyo, Japan
来源：
2014 4TH JOINT WORKSHOP ON HANDS-FREE SPEECH COMMUNICATION AND MICROPHONE ARRAYS (HSCMA) | 2014年
关键词：
NMF; multichannel signal separation; music signal processing; spectrogram restoration; optimal divergence; AUDIO SOURCE SEPARATION; MIXTURES;
D O I：
暂无
中图分类号：
O42 [声学];
学科分类号：
070206 ; 082403 ;
摘要：
In this paper, we address an optimization issue for the divergence in supervised nonnegative matrix factorization with spectrogram restoration, which has been proposed for addressing multichannel signal separation. This method separates non-target components and reconstructs some missing data caused by preceding spatial clustering via supervised basis extrapolation. In our previous study, we only used a limited type of divergence, whereas the divergence selection is essential. Therefore, we extend this method to a more generalized form and give a theoretical analysis of the divergence optimization, where we reveal the trade-off between separation and extrapolation abilities.
引用
收藏
页码：92 / 96
页数：5
相关论文
共 43 条
[31]   SCORE INFORMED AUDIO SOURCE SEPARATION USING CONSTRAINED NONNEGATIVE MATRIX FACTORIZATION AND SCORE SYNTHESIS [J].
Fritsch, Joachim ;
Plumbley, Mark D. .
2013 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2013, :888-891
[32]   Joint-diagonalizability-constrained multichannel nonnegative matrix factorization based on time-variant multivariate complex sub-Gaussian distribution [J].
Kamo, Keigo ;
Mitsui, Yoshiki ;
Kubo, Yuki ;
Takamune, Norihiro ;
Kitamura, Daichi ;
Saruwatari, Hiroshi ;
Takahashi, Yu ;
Kondo, Kazunobu .
SIGNAL PROCESSING, 2021, 188
[33]   Single-Channel Signal Separation Using Spectral Basis Correlation with Sparse Nonnegative Tensor Factorization [J].
Parathai, P. ;
Tengtrairat, N. ;
Woo, W. L. ;
Gao, Bin .
CIRCUITS SYSTEMS AND SIGNAL PROCESSING, 2019, 38 (12) :5786-5816
[34]   REGULARIZED FAST MULTICHANNEL NONNEGATIVE MATRIX FACTORIZATION WITH ILRMA-BASED PRIOR DISTRIBUTION OF JOINT-DIAGONALIZATION PROCESS [J].
Kamo, Keigo ;
Kubo, Yuki ;
Takamune, Norihiro ;
Kitamura, Daichi ;
Saruwatari, Hiroshi ;
Takahashi, Yu ;
Kondo, Kazunobu .
2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, :606-610
[35]   JOINT SEPARATION AND DEREVERBERATION OF REVERBERANT MIXTURES WITH DETERMINED MULTICHANNEL NON-NEGATIVE MATRIX FACTORIZATION [J].
Kagami, Hideaki ;
Kameoka, Hirokazu ;
Yukawa, Masahiro .
2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018, :31-35
[36]   Interference Separation Using Improved Nonnegative Matrix Factorization for Odor Reproduction Analysis in Mass Spectrum Space [J].
Prasetyawan, Dani ;
Zhao, Hanqing ;
Nakamoto, Takamichi .
2024 IEEE INTERNATIONAL SYMPOSIUM ON OLFACTION AND ELECTRONIC NOSE, ISOEN, 2024,
[37]   Joint-Diagonalizability-Constrained Multichannel Nonnegative Matrix Factorization Based on Multivariate Complex Student's t-distribution [J].
Kamo, Keigo ;
Kubo, Yuki ;
Takamune, Norihiro ;
Kitamura, Daichi ;
Saruwatari, Hiroshi ;
Takahashi, Yu ;
Kondo, Kazunobu .
2020 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC), 2020, :869-874
[38]   On-Line Nonnegative Matrix Factorization Method Using Recursive Least Squares for Acoustic Signal Processing Systems [J].
Lee, Seokjin ;
Park, Sang Ha ;
Sung, Koeng-Mo .
IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 2011, E94A (10) :2022-2026
[39]   RLS-Based On-Line Sparse Nonnegative Matrix Factorization Method for Acoustic Signal Processing Systems [J].
Lee, Seokjin .
IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 2013, E96A (05) :980-985
[40]   Speech source separation avoiding initial value dependency by cepstral-basis-decom-posed nonnegative matrix factorization [J].
Oshima, Fuga ;
Nakayama, Masashi .
ACOUSTICAL SCIENCE AND TECHNOLOGY, 2023, 44 (02) :137-140
12345