Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

1 Citation (Scopus)
178 Downloads (Pure)

Resumé

In this paper, we propose a method for finding the number of sources and their parameters from stereophonic mixtures. The method is based on clustering of narrowband interaural level and time differences for an unknown number of sources and uses an optimal segmentation on which the clustering is based. The parameter distribution, for both individual seg- ments and across segments that comprise the entire signal, is modelled as a Gaussian mixture. For each segment parame- ters are estimated using a minimum description length algo- rithm for mixtures based on the expectation-maximization al- gorithm. The generalized variance and degree of membership of the Gaussian components across segments is used as a ba- sis for the proposed selection of clusters amongst candidates. Simulations on synthetic and real audio shows promising re- sults for source parameter estimation and number of sources estimated across segments. The optimal segmentation shows an improvement for parameter estimation success rate, com- pared to the uniform segmentation.
OriginalsprogEngelsk
TitelIEEE International Conference on Acoustics, Speech and Signal Processing
Antal sider5
ForlagIEEE
Publikationsdato10 sep. 2018
Sider426-430
Artikelnummer8462522
ISBN (Trykt)978-1-5386-4659-5
ISBN (Elektronisk)978-1-5386-4658-8
DOI
StatusUdgivet - 10 sep. 2018
Begivenhed2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) - Calgary, Canada
Varighed: 15 apr. 201820 apr. 2018
https://2018.ieeeicassp.org/

Konference

Konference2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
LandCanada
ByCalgary
Periode15/04/201820/04/2018
Internetadresse
NavnI E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings
ISSN1520-6149

Fingerprint

Parameter estimation

Citer dette

Hjerrild, J. M., & Christensen, M. G. (2018). Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures. I IEEE International Conference on Acoustics, Speech and Signal Processing (s. 426-430). [8462522] IEEE. I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings https://doi.org/10.1109/ICASSP.2018.8462522
Hjerrild, Jacob Møller ; Christensen, Mads Græsbøll. / Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures. IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, 2018. s. 426-430 (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings).
@inproceedings{e1e44d7b2eb04ddb99b9340a75522ed4,
title = "Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures",
abstract = "In this paper, we propose a method for finding the number of sources and their parameters from stereophonic mixtures. The method is based on clustering of narrowband interaural level and time differences for an unknown number of sources and uses an optimal segmentation on which the clustering is based. The parameter distribution, for both individual seg- ments and across segments that comprise the entire signal, is modelled as a Gaussian mixture. For each segment parame- ters are estimated using a minimum description length algo- rithm for mixtures based on the expectation-maximization al- gorithm. The generalized variance and degree of membership of the Gaussian components across segments is used as a ba- sis for the proposed selection of clusters amongst candidates. Simulations on synthetic and real audio shows promising re- sults for source parameter estimation and number of sources estimated across segments. The optimal segmentation shows an improvement for parameter estimation success rate, com- pared to the uniform segmentation.",
keywords = "Audio analysis, Audio clustering, Multi-channel processing, Signal segmentation, Source localisation",
author = "Hjerrild, {Jacob M{\o}ller} and Christensen, {Mads Gr{\ae}sb{\o}ll}",
year = "2018",
month = "9",
day = "10",
doi = "10.1109/ICASSP.2018.8462522",
language = "English",
isbn = "978-1-5386-4659-5",
pages = "426--430",
booktitle = "IEEE International Conference on Acoustics, Speech and Signal Processing",
publisher = "IEEE",
address = "United States",

}

Hjerrild, JM & Christensen, MG 2018, Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures. i IEEE International Conference on Acoustics, Speech and Signal Processing., 8462522, IEEE, I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings, s. 426-430, Calgary, Canada, 15/04/2018. https://doi.org/10.1109/ICASSP.2018.8462522

Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures. / Hjerrild, Jacob Møller; Christensen, Mads Græsbøll.

IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, 2018. s. 426-430 8462522 (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings).

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures

AU - Hjerrild, Jacob Møller

AU - Christensen, Mads Græsbøll

PY - 2018/9/10

Y1 - 2018/9/10

N2 - In this paper, we propose a method for finding the number of sources and their parameters from stereophonic mixtures. The method is based on clustering of narrowband interaural level and time differences for an unknown number of sources and uses an optimal segmentation on which the clustering is based. The parameter distribution, for both individual seg- ments and across segments that comprise the entire signal, is modelled as a Gaussian mixture. For each segment parame- ters are estimated using a minimum description length algo- rithm for mixtures based on the expectation-maximization al- gorithm. The generalized variance and degree of membership of the Gaussian components across segments is used as a ba- sis for the proposed selection of clusters amongst candidates. Simulations on synthetic and real audio shows promising re- sults for source parameter estimation and number of sources estimated across segments. The optimal segmentation shows an improvement for parameter estimation success rate, com- pared to the uniform segmentation.

AB - In this paper, we propose a method for finding the number of sources and their parameters from stereophonic mixtures. The method is based on clustering of narrowband interaural level and time differences for an unknown number of sources and uses an optimal segmentation on which the clustering is based. The parameter distribution, for both individual seg- ments and across segments that comprise the entire signal, is modelled as a Gaussian mixture. For each segment parame- ters are estimated using a minimum description length algo- rithm for mixtures based on the expectation-maximization al- gorithm. The generalized variance and degree of membership of the Gaussian components across segments is used as a ba- sis for the proposed selection of clusters amongst candidates. Simulations on synthetic and real audio shows promising re- sults for source parameter estimation and number of sources estimated across segments. The optimal segmentation shows an improvement for parameter estimation success rate, com- pared to the uniform segmentation.

KW - Audio analysis

KW - Audio clustering

KW - Multi-channel processing

KW - Signal segmentation

KW - Source localisation

UR - http://www.scopus.com/inward/record.url?scp=85054255774&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2018.8462522

DO - 10.1109/ICASSP.2018.8462522

M3 - Article in proceeding

SN - 978-1-5386-4659-5

SP - 426

EP - 430

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

PB - IEEE

ER -

Hjerrild JM, Christensen MG. Estimation of Source Panning Parameters and Segmentation of Stereophonic Mixtures. I IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE. 2018. s. 426-430. 8462522. (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings). https://doi.org/10.1109/ICASSP.2018.8462522