Localization of multiple simultaneously active speakers in an acoustic sensor network

Andreas Brendel; Sharon Gannot; Walter Kellermann

doi:10.1109/sam.2018.8448802

Localization of multiple simultaneously active speakers in an acoustic sensor network

Andreas Brendel
, Sharon Gannot
, Walter Kellermann

Bar-Ilan University - The Alexander Kofkin Faculty of Engineering

Friedrich-Alexander University Erlangen-Nürnberg

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

6 Scopus citations

Abstract

This paper addresses the localization of an unknown number of acoustic sources in an enclosure. We extend a well established algorithm for localization of acoustic sources, which is based on the Expectation Maximization (EM) algorithm for clustering of phase differences by a Gaussian mixture model. Supporting a more appropriate probabilistic model for spherical data such as direction of arrival or phase differences, the von Mises distribution is used to derive a localization algorithm for multiple simultaneously active sources. Experiments with simulated room impulse responses confirm the superiority of the proposed algorithm to the existing method in terms of localization performance.

Original language	English
Title of host publication	2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018
Publisher	IEEE Computer Society
Pages	450-454
Number of pages	5
ISBN (Print)	9781538647523
DOIs	https://doi.org/10.1109/sam.2018.8448802
State	Published - 27 Aug 2018
Event	10th IEEE Sensor Array and Multichannel Signal Processing Workshop, SAM 2018 - Sheffield, United Kingdom Duration: 8 Jul 2018 → 11 Jul 2018

Publication series

Name	Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop
Volume	2018-July
ISSN (Electronic)	2151-870X

Conference

Conference	10th IEEE Sensor Array and Multichannel Signal Processing Workshop, SAM 2018
Country/Territory	United Kingdom
City	Sheffield
Period	8/07/18 → 11/07/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Funding

ACKNOWLEDGMENT This work was supported by DFG under contract no <Ke890/10-1> within the Research Unit FOR2457 ”Acoustic Sensor Networks” Fig. 2: Box plot of the localization error as defined in (17) for the method [15] (abbreviated with GMM) and the proposed method (abbreviated with vM). The number after the abbreviation gives the number of microphone pairs.

Funders	Funder number
Deutsche Forschungsgemeinschaft	<Ke890/10-1

Access to Document

10.1109/sam.2018.8448802

Cite this

Brendel, A., Gannot, S., & Kellermann, W. (2018). Localization of multiple simultaneously active speakers in an acoustic sensor network. In 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018 (pp. 450-454). Article 8448802 (Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop; Vol. 2018-July). IEEE Computer Society. https://doi.org/10.1109/sam.2018.8448802

@inproceedings{e225368dea94401b9bdc79d2151783c3,

title = "Localization of multiple simultaneously active speakers in an acoustic sensor network",

abstract = "This paper addresses the localization of an unknown number of acoustic sources in an enclosure. We extend a well established algorithm for localization of acoustic sources, which is based on the Expectation Maximization (EM) algorithm for clustering of phase differences by a Gaussian mixture model. Supporting a more appropriate probabilistic model for spherical data such as direction of arrival or phase differences, the von Mises distribution is used to derive a localization algorithm for multiple simultaneously active sources. Experiments with simulated room impulse responses confirm the superiority of the proposed algorithm to the existing method in terms of localization performance.",

author = "Andreas Brendel and Sharon Gannot and Walter Kellermann",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 10th IEEE Sensor Array and Multichannel Signal Processing Workshop, SAM 2018 ; Conference date: 08-07-2018 Through 11-07-2018",

year = "2018",

month = aug,

day = "27",

doi = "10.1109/sam.2018.8448802",

language = "אנגלית",

isbn = "9781538647523",

series = "Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop",

publisher = "IEEE Computer Society",

pages = "450--454",

booktitle = "2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018",

address = "ארצות הברית",

}

Brendel, A, Gannot, S & Kellermann, W 2018, Localization of multiple simultaneously active speakers in an acoustic sensor network. in 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018., 8448802, Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop, vol. 2018-July, IEEE Computer Society, pp. 450-454, 10th IEEE Sensor Array and Multichannel Signal Processing Workshop, SAM 2018, Sheffield, United Kingdom, 8/07/18. https://doi.org/10.1109/sam.2018.8448802

Localization of multiple simultaneously active speakers in an acoustic sensor network. / Brendel, Andreas; Gannot, Sharon; Kellermann, Walter.
2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018. IEEE Computer Society, 2018. p. 450-454 8448802 (Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop; Vol. 2018-July).

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

TY - GEN

T1 - Localization of multiple simultaneously active speakers in an acoustic sensor network

AU - Brendel, Andreas

AU - Gannot, Sharon

AU - Kellermann, Walter

PY - 2018/8/27

Y1 - 2018/8/27

N2 - This paper addresses the localization of an unknown number of acoustic sources in an enclosure. We extend a well established algorithm for localization of acoustic sources, which is based on the Expectation Maximization (EM) algorithm for clustering of phase differences by a Gaussian mixture model. Supporting a more appropriate probabilistic model for spherical data such as direction of arrival or phase differences, the von Mises distribution is used to derive a localization algorithm for multiple simultaneously active sources. Experiments with simulated room impulse responses confirm the superiority of the proposed algorithm to the existing method in terms of localization performance.

AB - This paper addresses the localization of an unknown number of acoustic sources in an enclosure. We extend a well established algorithm for localization of acoustic sources, which is based on the Expectation Maximization (EM) algorithm for clustering of phase differences by a Gaussian mixture model. Supporting a more appropriate probabilistic model for spherical data such as direction of arrival or phase differences, the von Mises distribution is used to derive a localization algorithm for multiple simultaneously active sources. Experiments with simulated room impulse responses confirm the superiority of the proposed algorithm to the existing method in terms of localization performance.

UR - https://www.scopus.com/pages/publications/85053640604

U2 - 10.1109/sam.2018.8448802

DO - 10.1109/sam.2018.8448802

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85053640604

SN - 9781538647523

T3 - Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop

SP - 450

EP - 454

BT - 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018

PB - IEEE Computer Society

T2 - 10th IEEE Sensor Array and Multichannel Signal Processing Workshop, SAM 2018

Y2 - 8 July 2018 through 11 July 2018

ER -

Brendel A, Gannot S, Kellermann W. Localization of multiple simultaneously active speakers in an acoustic sensor network. In 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018. IEEE Computer Society. 2018. p. 450-454. 8448802. (Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop). doi: 10.1109/sam.2018.8448802

Localization of multiple simultaneously active speakers in an acoustic sensor network

Abstract

Publication series

Conference

Bibliographical note

Funding

Access to Document

Other files and links

Fingerprint

Cite this