On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation

Matt Atcheson, I. Jafari, Roberto Togneri, S.E. Nordholm

Research output: Chapter in Book/Conference paperConference paper

3 Citations (Scopus)

Abstract

In this paper we propose to incorporate contextual time-frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consider the contextual information of each time-frequency slot. Motivated by the homogenous behavior of speech signals, we modify the fuzzy c-means clustering to bias the results in favor of cluster membership homogeneity within localized neighborhoods in the time-frequency space. Experimental evaluations in both simulated and real-world underdetermined environments demonstrate improvement in source separation performance over previous clustering approaches. © 2014 IEEE.
Original languageEnglish
Title of host publicationIEEE International Conference on Acoustics, Speech and Signal Processing
Place of PublicationFlorence, Italy
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages2114-2118
VolumeNA
ISBN (Print)9781479928927
DOIs
Publication statusPublished - 2014
Event2014 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2014 - Italy, Florence, Italy
Duration: 4 May 20149 May 2014
Conference number: 106632

Conference

Conference2014 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2014
Abbreviated titleICASSP 2014
CountryItaly
CityFlorence
Period4/05/149/05/14

Fingerprint

Source separation
Blind source separation
Masks

Cite this

Atcheson, M., Jafari, I., Togneri, R., & Nordholm, S. E. (2014). On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation. In IEEE International Conference on Acoustics, Speech and Signal Processing (Vol. NA, pp. 2114-2118). Florence, Italy: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICASSP.2014.6853972
Atcheson, Matt ; Jafari, I. ; Togneri, Roberto ; Nordholm, S.E. / On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation. IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. NA Florence, Italy : IEEE, Institute of Electrical and Electronics Engineers, 2014. pp. 2114-2118
@inproceedings{d8611ed88e3649268d306b36fc365f2c,
title = "On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation",
abstract = "In this paper we propose to incorporate contextual time-frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consider the contextual information of each time-frequency slot. Motivated by the homogenous behavior of speech signals, we modify the fuzzy c-means clustering to bias the results in favor of cluster membership homogeneity within localized neighborhoods in the time-frequency space. Experimental evaluations in both simulated and real-world underdetermined environments demonstrate improvement in source separation performance over previous clustering approaches. {\circledC} 2014 IEEE.",
author = "Matt Atcheson and I. Jafari and Roberto Togneri and S.E. Nordholm",
year = "2014",
doi = "10.1109/ICASSP.2014.6853972",
language = "English",
isbn = "9781479928927",
volume = "NA",
pages = "2114--2118",
booktitle = "IEEE International Conference on Acoustics, Speech and Signal Processing",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Atcheson, M, Jafari, I, Togneri, R & Nordholm, SE 2014, On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation. in IEEE International Conference on Acoustics, Speech and Signal Processing. vol. NA, IEEE, Institute of Electrical and Electronics Engineers, Florence, Italy, pp. 2114-2118, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2014, Florence, Italy, 4/05/14. https://doi.org/10.1109/ICASSP.2014.6853972

On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation. / Atcheson, Matt; Jafari, I.; Togneri, Roberto; Nordholm, S.E.

IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. NA Florence, Italy : IEEE, Institute of Electrical and Electronics Engineers, 2014. p. 2114-2118.

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation

AU - Atcheson, Matt

AU - Jafari, I.

AU - Togneri, Roberto

AU - Nordholm, S.E.

PY - 2014

Y1 - 2014

N2 - In this paper we propose to incorporate contextual time-frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consider the contextual information of each time-frequency slot. Motivated by the homogenous behavior of speech signals, we modify the fuzzy c-means clustering to bias the results in favor of cluster membership homogeneity within localized neighborhoods in the time-frequency space. Experimental evaluations in both simulated and real-world underdetermined environments demonstrate improvement in source separation performance over previous clustering approaches. © 2014 IEEE.

AB - In this paper we propose to incorporate contextual time-frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consider the contextual information of each time-frequency slot. Motivated by the homogenous behavior of speech signals, we modify the fuzzy c-means clustering to bias the results in favor of cluster membership homogeneity within localized neighborhoods in the time-frequency space. Experimental evaluations in both simulated and real-world underdetermined environments demonstrate improvement in source separation performance over previous clustering approaches. © 2014 IEEE.

U2 - 10.1109/ICASSP.2014.6853972

DO - 10.1109/ICASSP.2014.6853972

M3 - Conference paper

SN - 9781479928927

VL - NA

SP - 2114

EP - 2118

BT - IEEE International Conference on Acoustics, Speech and Signal Processing

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Florence, Italy

ER -

Atcheson M, Jafari I, Togneri R, Nordholm SE. On the use of contextual time-frequency information for full-band clustering-based convolutive blind source separation. In IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. NA. Florence, Italy: IEEE, Institute of Electrical and Electronics Engineers. 2014. p. 2114-2118 https://doi.org/10.1109/ICASSP.2014.6853972