Batch and adaptive PARAFAC-based blind separation of convolutive speech mixtures

Dimitri Nion; Kleanthis N. Mokios; Nicholas D. Sidiropoulos; Alexandros Potamianos

doi:10.1109/TASL.2009.2031694

Batch and adaptive PARAFAC-based blind separation of convolutive speech mixtures

Dimitri Nion, Kleanthis N. Mokios, Nicholas D. Sidiropoulos, Alexandros Potamianos

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

85 Scopus citations

Abstract

We present a frequency-domain technique based on PARAllel FACtor (PARAFAC) analysis that performs multichannel blind source separation (BSS) of convolutive speech mixtures. PARAFAC algorithms are combined with a dimensionality reduction step to significantly reduce computational complexity. The identifiability potential of PARAFAC is exploited to derive a BSS algorithm for the under-determined case (more speakers than microphones), combining PARAFAC analysis with time-varying Capon beamforming. Finally, a low-complexity adaptive version of the BSS algorithm is proposed that can track changes in the mixing environment. Extensive experiments with realistic and measured data corroborate our claims, including the under-determined case. Signal-to-interference ratio improvements of up to 6 dB are shown compared to state-of-the-art BSS algorithms, at an order of magnitude lower computational complexity.

Original language	English (US)
Article number	5233821
Pages (from-to)	1193-1207
Number of pages	15
Journal	IEEE Transactions on Audio, Speech and Language Processing
Volume	18
Issue number	6
DOIs	https://doi.org/10.1109/TASL.2009.2031694
State	Published - 2010

Bibliographical note

Funding Information:
Manuscript received June 24, 2008; revised July 31, 2009. First published September 09, 2009; current version published July 14, 2010. The work of D. Nion was supported by a postdoctoral grant from the Délégation Générale pour l’Armement (DGA) via ETIS Lab., UMR 8051 (ENSEA, CNRS, University of Cergy-Pontoise), France. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Jingdong Chen.

Keywords

Adaptive separation
PARAllel FACtor (PARAFAC)
blind speech separation
joint diagonalization
permutation ambiguity
underdetermined case

Publisher link

10.1109/TASL.2009.2031694

Find It

Find It at U of M Libraries

Cite this

@article{16348897db6343e6bc3d9d0cb063d6b2,

title = "Batch and adaptive PARAFAC-based blind separation of convolutive speech mixtures",

abstract = "We present a frequency-domain technique based on PARAllel FACtor (PARAFAC) analysis that performs multichannel blind source separation (BSS) of convolutive speech mixtures. PARAFAC algorithms are combined with a dimensionality reduction step to significantly reduce computational complexity. The identifiability potential of PARAFAC is exploited to derive a BSS algorithm for the under-determined case (more speakers than microphones), combining PARAFAC analysis with time-varying Capon beamforming. Finally, a low-complexity adaptive version of the BSS algorithm is proposed that can track changes in the mixing environment. Extensive experiments with realistic and measured data corroborate our claims, including the under-determined case. Signal-to-interference ratio improvements of up to 6 dB are shown compared to state-of-the-art BSS algorithms, at an order of magnitude lower computational complexity.",

keywords = "Adaptive separation, PARAllel FACtor (PARAFAC), blind speech separation, joint diagonalization, permutation ambiguity, underdetermined case",

author = "Dimitri Nion and Mokios, {Kleanthis N.} and Sidiropoulos, {Nicholas D.} and Alexandros Potamianos",

note = "Funding Information: Manuscript received June 24, 2008; revised July 31, 2009. First published September 09, 2009; current version published July 14, 2010. The work of D. Nion was supported by a postdoctoral grant from the D{\'e}l{\'e}gation G{\'e}n{\'e}rale pour l{\textquoteright}Armement (DGA) via ETIS Lab., UMR 8051 (ENSEA, CNRS, University of Cergy-Pontoise), France. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Jingdong Chen.",

year = "2010",

doi = "10.1109/TASL.2009.2031694",

language = "English (US)",

volume = "18",

pages = "1193--1207",

journal = "IEEE Transactions on Speech and Audio Processing",

issn = "1558-7916",

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

number = "6",

}

TY - JOUR

T1 - Batch and adaptive PARAFAC-based blind separation of convolutive speech mixtures

AU - Nion, Dimitri

AU - Mokios, Kleanthis N.

AU - Sidiropoulos, Nicholas D.

AU - Potamianos, Alexandros

N1 - Funding Information: Manuscript received June 24, 2008; revised July 31, 2009. First published September 09, 2009; current version published July 14, 2010. The work of D. Nion was supported by a postdoctoral grant from the Délégation Générale pour l’Armement (DGA) via ETIS Lab., UMR 8051 (ENSEA, CNRS, University of Cergy-Pontoise), France. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Jingdong Chen.

PY - 2010

Y1 - 2010

N2 - We present a frequency-domain technique based on PARAllel FACtor (PARAFAC) analysis that performs multichannel blind source separation (BSS) of convolutive speech mixtures. PARAFAC algorithms are combined with a dimensionality reduction step to significantly reduce computational complexity. The identifiability potential of PARAFAC is exploited to derive a BSS algorithm for the under-determined case (more speakers than microphones), combining PARAFAC analysis with time-varying Capon beamforming. Finally, a low-complexity adaptive version of the BSS algorithm is proposed that can track changes in the mixing environment. Extensive experiments with realistic and measured data corroborate our claims, including the under-determined case. Signal-to-interference ratio improvements of up to 6 dB are shown compared to state-of-the-art BSS algorithms, at an order of magnitude lower computational complexity.

AB - We present a frequency-domain technique based on PARAllel FACtor (PARAFAC) analysis that performs multichannel blind source separation (BSS) of convolutive speech mixtures. PARAFAC algorithms are combined with a dimensionality reduction step to significantly reduce computational complexity. The identifiability potential of PARAFAC is exploited to derive a BSS algorithm for the under-determined case (more speakers than microphones), combining PARAFAC analysis with time-varying Capon beamforming. Finally, a low-complexity adaptive version of the BSS algorithm is proposed that can track changes in the mixing environment. Extensive experiments with realistic and measured data corroborate our claims, including the under-determined case. Signal-to-interference ratio improvements of up to 6 dB are shown compared to state-of-the-art BSS algorithms, at an order of magnitude lower computational complexity.

KW - Adaptive separation

KW - PARAllel FACtor (PARAFAC)

KW - blind speech separation

KW - joint diagonalization

KW - permutation ambiguity

KW - underdetermined case

UR - http://www.scopus.com/inward/record.url?scp=77955837319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955837319&partnerID=8YFLogxK

U2 - 10.1109/TASL.2009.2031694

DO - 10.1109/TASL.2009.2031694

M3 - Article

AN - SCOPUS:77955837319

VL - 18

SP - 1193

EP - 1207

JO - IEEE Transactions on Speech and Audio Processing

JF - IEEE Transactions on Speech and Audio Processing

SN - 1558-7916

IS - 6

M1 - 5233821

ER -

Batch and adaptive PARAFAC-based blind separation of convolutive speech mixtures

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this