Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity

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

4 Citationer (Scopus)
402 Downloads (Pure)

Resumé

In this paper, we consider the problem of multi-pitch estimation and tracking of an unknown number of harmonic audio sources. The regularized least-squares is a solution for simultaneous sparse source selection and parameter estimation. Exploiting block sparsity, the method allows for reliable tracking of the found sources, without posing detailed a priori assumptions of the number of harmonics for each source. The method incorporates a Bayesian prior and assigns data-dependent regularization coefficients to efficiently incorporate both earlier and future data blocks in the tracking of estimates. In comparison with fix regularization coefficients, the simulation results, using both real and synthetic audio signals, confirm the performance of the proposed method.
OriginalsprogEngelsk
Titel2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015)
ForlagIEEE
Publikationsdato2015
Sider16-20
ISBN (Trykt)978-0-9928626-3-3
DOI
StatusUdgivet - 2015
BegivenhedEuropean Signal Processing Conference (EUSIPCO) - Nice, Frankrig
Varighed: 31 aug. 20154 sep. 2015

Konference

KonferenceEuropean Signal Processing Conference (EUSIPCO)
LandFrankrig
ByNice
Periode31/08/201504/09/2015
NavnProceedings of the European Signal Processing Conference
ISSN2076-1465

Fingerprint

Parameter estimation

Citer dette

Karimian-Azari, S., Jakobsson, A., Jensen, J. R., & Christensen, M. G. (2015). Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity. I 2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015) (s. 16-20). IEEE. Proceedings of the European Signal Processing Conference https://doi.org/10.1109/EUSIPCO.2015.7362336
Karimian-Azari, Sam ; Jakobsson, Andreas ; Jensen, Jesper Rindom ; Christensen, Mads Græsbøll. / Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity. 2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015) . IEEE, 2015. s. 16-20 (Proceedings of the European Signal Processing Conference).
@inproceedings{acb43ee33f8d47a9997d7f25e13cf8aa,
title = "Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity",
abstract = "In this paper, we consider the problem of multi-pitch estimation and tracking of an unknown number of harmonic audio sources. The regularized least-squares is a solution for simultaneous sparse source selection and parameter estimation. Exploiting block sparsity, the method allows for reliable tracking of the found sources, without posing detailed a priori assumptions of the number of harmonics for each source. The method incorporates a Bayesian prior and assigns data-dependent regularization coefficients to efficiently incorporate both earlier and future data blocks in the tracking of estimates. In comparison with fix regularization coefficients, the simulation results, using both real and synthetic audio signals, confirm the performance of the proposed method.",
keywords = "Multi-pitch estimation, tracking, harmonic signal, regularized least-squares, sparsity",
author = "Sam Karimian-Azari and Andreas Jakobsson and Jensen, {Jesper Rindom} and Christensen, {Mads Gr{\ae}sb{\o}ll}",
year = "2015",
doi = "10.1109/EUSIPCO.2015.7362336",
language = "English",
isbn = "978-0-9928626-3-3",
pages = "16--20",
booktitle = "2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015)",
publisher = "IEEE",
address = "United States",

}

Karimian-Azari, S, Jakobsson, A, Jensen, JR & Christensen, MG 2015, Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity. i 2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015) . IEEE, Proceedings of the European Signal Processing Conference, s. 16-20, Nice, Frankrig, 31/08/2015. https://doi.org/10.1109/EUSIPCO.2015.7362336

Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity. / Karimian-Azari, Sam; Jakobsson, Andreas; Jensen, Jesper Rindom; Christensen, Mads Græsbøll.

2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015) . IEEE, 2015. s. 16-20 (Proceedings of the European Signal Processing Conference).

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

TY - GEN

T1 - Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity

AU - Karimian-Azari, Sam

AU - Jakobsson, Andreas

AU - Jensen, Jesper Rindom

AU - Christensen, Mads Græsbøll

PY - 2015

Y1 - 2015

N2 - In this paper, we consider the problem of multi-pitch estimation and tracking of an unknown number of harmonic audio sources. The regularized least-squares is a solution for simultaneous sparse source selection and parameter estimation. Exploiting block sparsity, the method allows for reliable tracking of the found sources, without posing detailed a priori assumptions of the number of harmonics for each source. The method incorporates a Bayesian prior and assigns data-dependent regularization coefficients to efficiently incorporate both earlier and future data blocks in the tracking of estimates. In comparison with fix regularization coefficients, the simulation results, using both real and synthetic audio signals, confirm the performance of the proposed method.

AB - In this paper, we consider the problem of multi-pitch estimation and tracking of an unknown number of harmonic audio sources. The regularized least-squares is a solution for simultaneous sparse source selection and parameter estimation. Exploiting block sparsity, the method allows for reliable tracking of the found sources, without posing detailed a priori assumptions of the number of harmonics for each source. The method incorporates a Bayesian prior and assigns data-dependent regularization coefficients to efficiently incorporate both earlier and future data blocks in the tracking of estimates. In comparison with fix regularization coefficients, the simulation results, using both real and synthetic audio signals, confirm the performance of the proposed method.

KW - Multi-pitch estimation

KW - tracking

KW - harmonic signal

KW - regularized least-squares

KW - sparsity

U2 - 10.1109/EUSIPCO.2015.7362336

DO - 10.1109/EUSIPCO.2015.7362336

M3 - Article in proceeding

SN - 978-0-9928626-3-3

SP - 16

EP - 20

BT - 2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015)

PB - IEEE

ER -

Karimian-Azari S, Jakobsson A, Jensen JR, Christensen MG. Multi-Pitch Estimation and Tracking Using Bayesian Inference in Block Sparsity. I 2015 Proceedings of the 23rd European Signal Processing Conference (EUSIPCO 2015) . IEEE. 2015. s. 16-20. (Proceedings of the European Signal Processing Conference). https://doi.org/10.1109/EUSIPCO.2015.7362336