Consistent ICA: Determined BSS Meets Spectrogram Consistency

被引：12

作者：

Yatabe, Kohei ^{[1
]}

机构：

[1] Waseda Univ, Dept Intermedia Art & Sci, Tokyo 1698555, Japan

来源：

IEEE SIGNAL PROCESSING LETTERS | 2020年 / 27卷

关键词：

Spectrogram; Time-frequency analysis; Time-domain analysis; Matrix converters; Blind source separation; Smoothing methods; Linear source separation; multichannel acoustic signal processing; demixing filter estimation; independent component analysis (ICA); short-time Fourier transform; INDEPENDENT COMPONENT ANALYSIS; SOURCE SEPARATION; PERMUTATION PROBLEM; ALGORITHMS; MIXTURES;

D O I：

10.1109/LSP.2020.2996904

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Multichannel audio blind source separation (BSS) in the determined situation (the number of microphones is equal to that of the sources), or determined BSS, is performed by multichannel linear filtering in the time-frequency domain to handle the convolutive mixing process. Ordinarily, the filter treats each frequency independently, which causes the well-known permutation problem, i.e., the problem of how to align the frequency-wise filters so that each separated component is correctly assigned to the corresponding sources. In this paper, it is shown that the general property of the time-frequency-domain representation called spectrogram consistency can be an assistant for solving the permutation problem.

引用

页码：870 / 874

页数：5

共 34 条

[1]

Anemuller J., 2000, Second International Workshop on Independent Component Analysis and Blind Signal Separation. Proceedings, P215

[2]

Araki Shoko, 2012, Latent Variable Analysis and Signal Separation. Proceedings 10th International Conference, LVA/ICA 2012, P414, DOI 10.1007/978-3-642-28551-6_51

[3] A generalization of blind source separation algorithms for convolutive mixtures based on second-order statistics [J].