Multichannel NMF for source separation with ambisonic signals

Joonas Nikunen; Archontis Politis

doi:10.1109/IWAENC.2018.8521344

Multichannel NMF for source separation with ambisonic signals

Joonas Nikunen, Archontis Politis

Department of Signal Processing and Acoustics

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

1 Citation (Scopus)

Abstract

This paper proposes a novel method for separation of sound sources with ambisonic signals using multichannel non-negative matrix factorization (MNMF) for source spectrogram estimation. We present a novel frequency-independent spatial covariance matrix (SCM) model for spherical harmonic (SH) domain signals which makes the MNMF parameter estimation framework computationally feasible up to 3rd order SH signals. The evaluation is done with simulated SH domain mixtures by measuring the separation performance using objective criteria and comparing the proposed method against SH domain beamforming. The proposed method improves average separation performance over beamforming with post-filtering when using 1st and 2nd order SH signals while at higher orders performance among all tested methods is similar.

Original language	English
Title of host publication	16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings
Publisher	IEEE
Pages	251-255
Number of pages	5
ISBN (Electronic)	9781538681510
DOIs	https://doi.org/10.1109/IWAENC.2018.8521344
Publication status	Published - 2 Nov 2018
MoE publication type	A4 Article in a conference publication
Event	International Workshop on Acoustic Signal Enhancement - Tokyo, Japan Duration: 17 Sep 2018 → 20 Sep 2018 Conference number: 16

Workshop

Workshop	International Workshop on Acoustic Signal Enhancement
Abbreviated title	IWAENC
Country	Japan
City	Tokyo
Period	17/09/2018 → 20/09/2018

Keywords

Ambisonics
Multichannel NMF
Source separation

Access to Document

10.1109/IWAENC.2018.8521344

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Multichannel NMF for source separation with ambisonic signals'. Together they form a unique fingerprint.

Cite this

@inproceedings{2cb59b8ac3a24668955061f78d463a23,

title = "Multichannel NMF for source separation with ambisonic signals",

abstract = "This paper proposes a novel method for separation of sound sources with ambisonic signals using multichannel non-negative matrix factorization (MNMF) for source spectrogram estimation. We present a novel frequency-independent spatial covariance matrix (SCM) model for spherical harmonic (SH) domain signals which makes the MNMF parameter estimation framework computationally feasible up to 3rd order SH signals. The evaluation is done with simulated SH domain mixtures by measuring the separation performance using objective criteria and comparing the proposed method against SH domain beamforming. The proposed method improves average separation performance over beamforming with post-filtering when using 1st and 2nd order SH signals while at higher orders performance among all tested methods is similar.",

keywords = "Ambisonics, Multichannel NMF, Source separation",

author = "Joonas Nikunen and Archontis Politis",

year = "2018",

month = nov,

day = "2",

doi = "10.1109/IWAENC.2018.8521344",

language = "English",

pages = "251--255",

booktitle = "16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings",

publisher = "IEEE",

address = "United States",

note = "International Workshop on Acoustic Signal Enhancement, IWAENC ; Conference date: 17-09-2018 Through 20-09-2018",

}

TY - GEN

T1 - Multichannel NMF for source separation with ambisonic signals

AU - Nikunen, Joonas

AU - Politis, Archontis

N1 - Conference code: 16

PY - 2018/11/2

Y1 - 2018/11/2

N2 - This paper proposes a novel method for separation of sound sources with ambisonic signals using multichannel non-negative matrix factorization (MNMF) for source spectrogram estimation. We present a novel frequency-independent spatial covariance matrix (SCM) model for spherical harmonic (SH) domain signals which makes the MNMF parameter estimation framework computationally feasible up to 3rd order SH signals. The evaluation is done with simulated SH domain mixtures by measuring the separation performance using objective criteria and comparing the proposed method against SH domain beamforming. The proposed method improves average separation performance over beamforming with post-filtering when using 1st and 2nd order SH signals while at higher orders performance among all tested methods is similar.

AB - This paper proposes a novel method for separation of sound sources with ambisonic signals using multichannel non-negative matrix factorization (MNMF) for source spectrogram estimation. We present a novel frequency-independent spatial covariance matrix (SCM) model for spherical harmonic (SH) domain signals which makes the MNMF parameter estimation framework computationally feasible up to 3rd order SH signals. The evaluation is done with simulated SH domain mixtures by measuring the separation performance using objective criteria and comparing the proposed method against SH domain beamforming. The proposed method improves average separation performance over beamforming with post-filtering when using 1st and 2nd order SH signals while at higher orders performance among all tested methods is similar.

KW - Ambisonics

KW - Multichannel NMF

KW - Source separation

UR - http://www.scopus.com/inward/record.url?scp=85057417393&partnerID=8YFLogxK

U2 - 10.1109/IWAENC.2018.8521344

DO - 10.1109/IWAENC.2018.8521344

M3 - Conference contribution

AN - SCOPUS:85057417393

SP - 251

EP - 255

BT - 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings

PB - IEEE

T2 - International Workshop on Acoustic Signal Enhancement

Y2 - 17 September 2018 through 20 September 2018

ER -