Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach

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

2 Citationer (Scopus)
168 Downloads (Pure)

Resumé

In this paper, a method for multi-pitch estimation of single-channel mixtures of harmonic signals is presented. Using the method, it is possible to resolve amplitudes of overlapping harmonics, which is otherwise an ill-posed problem. The method is based on the extended invariance principle (EXIP), and a codebook consisting of realistic amplitude vectors. A nonlinear least squares (NLS) cost function is formed based on the observed signal and a parametric model of the signal, for a set of fundamental frequency candidates. For each of these, amplitude estimates are computed. The magnitudes of these estimates are quantized according to a codebook, and an updated cost function is used to estimate the fundamental frequencies of the sources. The performance of the proposed estimator is evaluated using synthetic and real mixtures, and the results show that the proposed method is able to estimate multiple pitches in a mixture of sources with overlapping harmonics.
OriginalsprogEngelsk
TitelProceedings of the 24th European Signal Processing Conference (EUSIPCO)
ForlagIEEE
Publikationsdato2016
Sider983-987
ISBN (Trykt)978-0-9928-6265-7
DOI
StatusUdgivet - 2016
Begivenhed24th European Signal Processing Conference - Hilton Budapest, Budapest, Ungarn
Varighed: 29 aug. 20162 sep. 2016
Konferencens nummer: 24

Konference

Konference24th European Signal Processing Conference
Nummer24
LokationHilton Budapest
LandUngarn
ByBudapest
Periode29/08/201602/09/2016
NavnProceedings of the European Signal Processing Conference (EUSIPCO)
ISSN2076-1465

Fingerprint

Audio recordings
Cost functions
Invariance

Citer dette

Hansen, M. W., Jensen, J. R., & Christensen, M. G. (2016). Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach. I Proceedings of the 24th European Signal Processing Conference (EUSIPCO) (s. 983-987). IEEE. Proceedings of the European Signal Processing Conference (EUSIPCO) https://doi.org/10.1109/EUSIPCO.2016.7760395
Hansen, Martin Weiss ; Jensen, Jesper Rindom ; Christensen, Mads Græsbøll. / Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach. Proceedings of the 24th European Signal Processing Conference (EUSIPCO). IEEE, 2016. s. 983-987 (Proceedings of the European Signal Processing Conference (EUSIPCO)).
@inproceedings{806d34eeaeb242b8a5a01c0e0c977184,
title = "Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach",
abstract = "In this paper, a method for multi-pitch estimation of single-channel mixtures of harmonic signals is presented. Using the method, it is possible to resolve amplitudes of overlapping harmonics, which is otherwise an ill-posed problem. The method is based on the extended invariance principle (EXIP), and a codebook consisting of realistic amplitude vectors. A nonlinear least squares (NLS) cost function is formed based on the observed signal and a parametric model of the signal, for a set of fundamental frequency candidates. For each of these, amplitude estimates are computed. The magnitudes of these estimates are quantized according to a codebook, and an updated cost function is used to estimate the fundamental frequencies of the sources. The performance of the proposed estimator is evaluated using synthetic and real mixtures, and the results show that the proposed method is able to estimate multiple pitches in a mixture of sources with overlapping harmonics.",
keywords = "multi-pitch estimation, amplitude estimation, vector quantization, music information retrieval",
author = "Hansen, {Martin Weiss} and Jensen, {Jesper Rindom} and Christensen, {Mads Gr{\ae}sb{\o}ll}",
year = "2016",
doi = "10.1109/EUSIPCO.2016.7760395",
language = "English",
isbn = "978-0-9928-6265-7",
series = "Proceedings of the European Signal Processing Conference (EUSIPCO)",
publisher = "IEEE",
pages = "983--987",
booktitle = "Proceedings of the 24th European Signal Processing Conference (EUSIPCO)",
address = "United States",

}

Hansen, MW, Jensen, JR & Christensen, MG 2016, Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach. i Proceedings of the 24th European Signal Processing Conference (EUSIPCO). IEEE, Proceedings of the European Signal Processing Conference (EUSIPCO), s. 983-987, Budapest, Ungarn, 29/08/2016. https://doi.org/10.1109/EUSIPCO.2016.7760395

Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach. / Hansen, Martin Weiss; Jensen, Jesper Rindom; Christensen, Mads Græsbøll.

Proceedings of the 24th European Signal Processing Conference (EUSIPCO). IEEE, 2016. s. 983-987 (Proceedings of the European Signal Processing Conference (EUSIPCO)).

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

TY - GEN

T1 - Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach

AU - Hansen, Martin Weiss

AU - Jensen, Jesper Rindom

AU - Christensen, Mads Græsbøll

PY - 2016

Y1 - 2016

N2 - In this paper, a method for multi-pitch estimation of single-channel mixtures of harmonic signals is presented. Using the method, it is possible to resolve amplitudes of overlapping harmonics, which is otherwise an ill-posed problem. The method is based on the extended invariance principle (EXIP), and a codebook consisting of realistic amplitude vectors. A nonlinear least squares (NLS) cost function is formed based on the observed signal and a parametric model of the signal, for a set of fundamental frequency candidates. For each of these, amplitude estimates are computed. The magnitudes of these estimates are quantized according to a codebook, and an updated cost function is used to estimate the fundamental frequencies of the sources. The performance of the proposed estimator is evaluated using synthetic and real mixtures, and the results show that the proposed method is able to estimate multiple pitches in a mixture of sources with overlapping harmonics.

AB - In this paper, a method for multi-pitch estimation of single-channel mixtures of harmonic signals is presented. Using the method, it is possible to resolve amplitudes of overlapping harmonics, which is otherwise an ill-posed problem. The method is based on the extended invariance principle (EXIP), and a codebook consisting of realistic amplitude vectors. A nonlinear least squares (NLS) cost function is formed based on the observed signal and a parametric model of the signal, for a set of fundamental frequency candidates. For each of these, amplitude estimates are computed. The magnitudes of these estimates are quantized according to a codebook, and an updated cost function is used to estimate the fundamental frequencies of the sources. The performance of the proposed estimator is evaluated using synthetic and real mixtures, and the results show that the proposed method is able to estimate multiple pitches in a mixture of sources with overlapping harmonics.

KW - multi-pitch estimation

KW - amplitude estimation

KW - vector quantization

KW - music information retrieval

U2 - 10.1109/EUSIPCO.2016.7760395

DO - 10.1109/EUSIPCO.2016.7760395

M3 - Article in proceeding

SN - 978-0-9928-6265-7

T3 - Proceedings of the European Signal Processing Conference (EUSIPCO)

SP - 983

EP - 987

BT - Proceedings of the 24th European Signal Processing Conference (EUSIPCO)

PB - IEEE

ER -

Hansen MW, Jensen JR, Christensen MG. Multi-Pitch Estimation of Audio Recordings Using a Codebook-Based Approach. I Proceedings of the 24th European Signal Processing Conference (EUSIPCO). IEEE. 2016. s. 983-987. (Proceedings of the European Signal Processing Conference (EUSIPCO)). https://doi.org/10.1109/EUSIPCO.2016.7760395