Cochlear pitch class profile for cover song identification

Ning Chen, J Stephen Downie, Hai Dong Xiao, Yu Zhu

Research output: Contribution to journalArticle

Abstract

Abstract Pitch class profile (PCP), which can represent the harmonic progression of a piece of music very well, is one of the widely used audio features for cover version identification. In this letter, we describe a novel procedure that enhances PCP by substantially boosting the degree of instrumental accompaniment invariance without degrading the feature's discriminative power. Our idea is based on the assumption that human ear can identify a cover of a pop song based on their singing voice quickly and easily. So, we combine two concepts from psychoacoustics: (i) time-varying loudness contour and (ii) critical band, which have been used in speech recognition field successfully, with the conventional PCP descriptor to enhance its discriminative power. Since the CPCPs aim at a representation of singing voice, they may also obtain improved performance (as compared to conventional PCPs) when applied to a cappella singing recordings. Experimental results demonstrate that the resulting PCP feature, called cochlear pitch class profile (CPCP), outperforms conventional PCP feature in the context of pop cover song identification application.

Original languageEnglish (US)
Article number5592
Pages (from-to)92-96
Number of pages5
JournalApplied Acoustics
Volume99
DOIs
StatePublished - Jul 10 2015

Fingerprint

profiles
psychoacoustics
loudness
music
speech recognition
ear
progressions
invariance
recording
harmonics

Keywords

  • Audio matching
  • Cochlear pitch class profile (CPCP)
  • Pop cover song identification

ASJC Scopus subject areas

  • Acoustics and Ultrasonics

Cite this

Cochlear pitch class profile for cover song identification. / Chen, Ning; Downie, J Stephen; Xiao, Hai Dong; Zhu, Yu.

In: Applied Acoustics, Vol. 99, 5592, 10.07.2015, p. 92-96.

Research output: Contribution to journalArticle

Chen, Ning ; Downie, J Stephen ; Xiao, Hai Dong ; Zhu, Yu. / Cochlear pitch class profile for cover song identification. In: Applied Acoustics. 2015 ; Vol. 99. pp. 92-96.
@article{664aca1e7ab94925bd4b5dc2c6040201,
title = "Cochlear pitch class profile for cover song identification",
abstract = "Abstract Pitch class profile (PCP), which can represent the harmonic progression of a piece of music very well, is one of the widely used audio features for cover version identification. In this letter, we describe a novel procedure that enhances PCP by substantially boosting the degree of instrumental accompaniment invariance without degrading the feature's discriminative power. Our idea is based on the assumption that human ear can identify a cover of a pop song based on their singing voice quickly and easily. So, we combine two concepts from psychoacoustics: (i) time-varying loudness contour and (ii) critical band, which have been used in speech recognition field successfully, with the conventional PCP descriptor to enhance its discriminative power. Since the CPCPs aim at a representation of singing voice, they may also obtain improved performance (as compared to conventional PCPs) when applied to a cappella singing recordings. Experimental results demonstrate that the resulting PCP feature, called cochlear pitch class profile (CPCP), outperforms conventional PCP feature in the context of pop cover song identification application.",
keywords = "Audio matching, Cochlear pitch class profile (CPCP), Pop cover song identification",
author = "Ning Chen and Downie, {J Stephen} and Xiao, {Hai Dong} and Yu Zhu",
year = "2015",
month = "7",
day = "10",
doi = "10.1016/j.apacoust.2015.06.003",
language = "English (US)",
volume = "99",
pages = "92--96",
journal = "Applied Acoustics",
issn = "0003-682X",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Cochlear pitch class profile for cover song identification

AU - Chen, Ning

AU - Downie, J Stephen

AU - Xiao, Hai Dong

AU - Zhu, Yu

PY - 2015/7/10

Y1 - 2015/7/10

N2 - Abstract Pitch class profile (PCP), which can represent the harmonic progression of a piece of music very well, is one of the widely used audio features for cover version identification. In this letter, we describe a novel procedure that enhances PCP by substantially boosting the degree of instrumental accompaniment invariance without degrading the feature's discriminative power. Our idea is based on the assumption that human ear can identify a cover of a pop song based on their singing voice quickly and easily. So, we combine two concepts from psychoacoustics: (i) time-varying loudness contour and (ii) critical band, which have been used in speech recognition field successfully, with the conventional PCP descriptor to enhance its discriminative power. Since the CPCPs aim at a representation of singing voice, they may also obtain improved performance (as compared to conventional PCPs) when applied to a cappella singing recordings. Experimental results demonstrate that the resulting PCP feature, called cochlear pitch class profile (CPCP), outperforms conventional PCP feature in the context of pop cover song identification application.

AB - Abstract Pitch class profile (PCP), which can represent the harmonic progression of a piece of music very well, is one of the widely used audio features for cover version identification. In this letter, we describe a novel procedure that enhances PCP by substantially boosting the degree of instrumental accompaniment invariance without degrading the feature's discriminative power. Our idea is based on the assumption that human ear can identify a cover of a pop song based on their singing voice quickly and easily. So, we combine two concepts from psychoacoustics: (i) time-varying loudness contour and (ii) critical band, which have been used in speech recognition field successfully, with the conventional PCP descriptor to enhance its discriminative power. Since the CPCPs aim at a representation of singing voice, they may also obtain improved performance (as compared to conventional PCPs) when applied to a cappella singing recordings. Experimental results demonstrate that the resulting PCP feature, called cochlear pitch class profile (CPCP), outperforms conventional PCP feature in the context of pop cover song identification application.

KW - Audio matching

KW - Cochlear pitch class profile (CPCP)

KW - Pop cover song identification

UR - http://www.scopus.com/inward/record.url?scp=84936858649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936858649&partnerID=8YFLogxK

U2 - 10.1016/j.apacoust.2015.06.003

DO - 10.1016/j.apacoust.2015.06.003

M3 - Article

AN - SCOPUS:84936858649

VL - 99

SP - 92

EP - 96

JO - Applied Acoustics

JF - Applied Acoustics

SN - 0003-682X

M1 - 5592

ER -