Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum

Kenji Ozawa; Masanori Morise; Shuichi Sakamoto; Kanji Watanabe

doi:10.1109/ICSAI48974.2019.9010477

Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum

Kenji Ozawa, Masanori Morise, Shuichi Sakamoto, Kanji Watanabe

Human Information Systems Division

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In our previous paper, we proposed a sound source separation method using the two-dimensional fast Fourier transform (2D FFT) of a spatio-temporal sound pressure distribution (STSPD) image that is composed from the outputs of a microphone array. In an STSPD image, vertical stripes are created for a target sound arriving from the perpendicular direction to the array; therefore, its spectral components are concentrated on the spatial direct current (DC) components in the 2D amplitude spectrum. In that study, we estimated the noise DC amplitudes using a deep neural network (DNN), then subtracted them from the observed spectrum to suppress the noise. However, the performance of noise suppression can be improved further. In this study, we estimate the noise DC components theoretically instead of empirically using a DNN. We improved the performance successfully.

Original language	English
Title of host publication	2019 6th International Conference on Systems and Informatics, ICSAI 2019
Editors	Wanqing Wu, Lipo Wang, Chunlei Ji, Niansheng Chen, Sun Qiang, Xiaoyong Song, Xin Wang
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1137-1142
Number of pages	6
ISBN (Electronic)	9781728152561
DOIs	https://doi.org/10.1109/ICSAI48974.2019.9010477
Publication status	Published - 2019 Nov
Event	6th International Conference on Systems and Informatics, ICSAI 2019 - Shanghai, China Duration: 2019 Nov 2 → 2019 Nov 4

Publication series

Name	2019 6th International Conference on Systems and Informatics, ICSAI 2019

Conference

Conference	6th International Conference on Systems and Informatics, ICSAI 2019
Country	China
City	Shanghai
Period	19/11/2 → 19/11/4

Keywords

Instantaneous estimation
Mathematical derivation
Spatio-temporal sound pressure distribution image
Spectral subtraction
Two-dimensional FFT

ASJC Scopus subject areas

Health Informatics
Computer Vision and Pattern Recognition
Information Systems
Control and Systems Engineering
Mechanical Engineering
Control and Optimization

Access to Document

10.1109/ICSAI48974.2019.9010477

Fingerprint Dive into the research topics of 'Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum'. Together they form a unique fingerprint.

Cite this

Ozawa, K., Morise, M., Sakamoto, S., & Watanabe, K. (2019). Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum. In W. Wu, L. Wang, C. Ji, N. Chen, S. Qiang, X. Song, & X. Wang (Eds.), 2019 6th International Conference on Systems and Informatics, ICSAI 2019 (pp. 1137-1142). [9010477] (2019 6th International Conference on Systems and Informatics, ICSAI 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICSAI48974.2019.9010477

Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum. / Ozawa, Kenji; Morise, Masanori; Sakamoto, Shuichi; Watanabe, Kanji.

2019 6th International Conference on Systems and Informatics, ICSAI 2019. ed. / Wanqing Wu; Lipo Wang; Chunlei Ji; Niansheng Chen; Sun Qiang; Xiaoyong Song; Xin Wang. Institute of Electrical and Electronics Engineers Inc., 2019. p. 1137-1142 9010477 (2019 6th International Conference on Systems and Informatics, ICSAI 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Ozawa, K, Morise, M, Sakamoto, S & Watanabe, K 2019, Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum. in W Wu, L Wang, C Ji, N Chen, S Qiang, X Song & X Wang (eds), 2019 6th International Conference on Systems and Informatics, ICSAI 2019., 9010477, 2019 6th International Conference on Systems and Informatics, ICSAI 2019, Institute of Electrical and Electronics Engineers Inc., pp. 1137-1142, 6th International Conference on Systems and Informatics, ICSAI 2019, Shanghai, China, 19/11/2. https://doi.org/10.1109/ICSAI48974.2019.9010477

Ozawa K, Morise M, Sakamoto S, Watanabe K. Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum. In Wu W, Wang L, Ji C, Chen N, Qiang S, Song X, Wang X, editors, 2019 6th International Conference on Systems and Informatics, ICSAI 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 1137-1142. 9010477. (2019 6th International Conference on Systems and Informatics, ICSAI 2019). https://doi.org/10.1109/ICSAI48974.2019.9010477

Ozawa, Kenji ; Morise, Masanori ; Sakamoto, Shuichi ; Watanabe, Kanji. / Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum. 2019 6th International Conference on Systems and Informatics, ICSAI 2019. editor / Wanqing Wu ; Lipo Wang ; Chunlei Ji ; Niansheng Chen ; Sun Qiang ; Xiaoyong Song ; Xin Wang. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 1137-1142 (2019 6th International Conference on Systems and Informatics, ICSAI 2019).

@inproceedings{6f301b92b18f4e19963ad3b2c248e751,

title = "Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum",

abstract = "In our previous paper, we proposed a sound source separation method using the two-dimensional fast Fourier transform (2D FFT) of a spatio-temporal sound pressure distribution (STSPD) image that is composed from the outputs of a microphone array. In an STSPD image, vertical stripes are created for a target sound arriving from the perpendicular direction to the array; therefore, its spectral components are concentrated on the spatial direct current (DC) components in the 2D amplitude spectrum. In that study, we estimated the noise DC amplitudes using a deep neural network (DNN), then subtracted them from the observed spectrum to suppress the noise. However, the performance of noise suppression can be improved further. In this study, we estimate the noise DC components theoretically instead of empirically using a DNN. We improved the performance successfully.",

keywords = "Instantaneous estimation, Mathematical derivation, Spatio-temporal sound pressure distribution image, Spectral subtraction, Two-dimensional FFT",

author = "Kenji Ozawa and Masanori Morise and Shuichi Sakamoto and Kanji Watanabe",

year = "2019",

month = nov,

doi = "10.1109/ICSAI48974.2019.9010477",

language = "English",

series = "2019 6th International Conference on Systems and Informatics, ICSAI 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1137--1142",

editor = "Wanqing Wu and Lipo Wang and Chunlei Ji and Niansheng Chen and Sun Qiang and Xiaoyong Song and Xin Wang",

booktitle = "2019 6th International Conference on Systems and Informatics, ICSAI 2019",

note = "6th International Conference on Systems and Informatics, ICSAI 2019 ; Conference date: 02-11-2019 Through 04-11-2019",

}

TY - GEN

T1 - Sound source separation by spectral subtraction based on instantaneous estimation of noise spectrum

AU - Ozawa, Kenji

AU - Morise, Masanori

AU - Sakamoto, Shuichi

AU - Watanabe, Kanji

PY - 2019/11

Y1 - 2019/11

N2 - In our previous paper, we proposed a sound source separation method using the two-dimensional fast Fourier transform (2D FFT) of a spatio-temporal sound pressure distribution (STSPD) image that is composed from the outputs of a microphone array. In an STSPD image, vertical stripes are created for a target sound arriving from the perpendicular direction to the array; therefore, its spectral components are concentrated on the spatial direct current (DC) components in the 2D amplitude spectrum. In that study, we estimated the noise DC amplitudes using a deep neural network (DNN), then subtracted them from the observed spectrum to suppress the noise. However, the performance of noise suppression can be improved further. In this study, we estimate the noise DC components theoretically instead of empirically using a DNN. We improved the performance successfully.

AB - In our previous paper, we proposed a sound source separation method using the two-dimensional fast Fourier transform (2D FFT) of a spatio-temporal sound pressure distribution (STSPD) image that is composed from the outputs of a microphone array. In an STSPD image, vertical stripes are created for a target sound arriving from the perpendicular direction to the array; therefore, its spectral components are concentrated on the spatial direct current (DC) components in the 2D amplitude spectrum. In that study, we estimated the noise DC amplitudes using a deep neural network (DNN), then subtracted them from the observed spectrum to suppress the noise. However, the performance of noise suppression can be improved further. In this study, we estimate the noise DC components theoretically instead of empirically using a DNN. We improved the performance successfully.

KW - Instantaneous estimation

KW - Mathematical derivation

KW - Spatio-temporal sound pressure distribution image

KW - Spectral subtraction

KW - Two-dimensional FFT

UR - http://www.scopus.com/inward/record.url?scp=85081958093&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85081958093&partnerID=8YFLogxK

U2 - 10.1109/ICSAI48974.2019.9010477

DO - 10.1109/ICSAI48974.2019.9010477

M3 - Conference contribution

AN - SCOPUS:85081958093

T3 - 2019 6th International Conference on Systems and Informatics, ICSAI 2019

SP - 1137

EP - 1142

BT - 2019 6th International Conference on Systems and Informatics, ICSAI 2019

A2 - Wu, Wanqing

A2 - Wang, Lipo

A2 - Ji, Chunlei

A2 - Chen, Niansheng

A2 - Qiang, Sun

A2 - Song, Xiaoyong

A2 - Wang, Xin

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 6th International Conference on Systems and Informatics, ICSAI 2019

Y2 - 2 November 2019 through 4 November 2019

ER -