Automatic sample recognition in hip-hop music based on non-negative matrix factorization

Jordan L. Whitney, Colby Leider

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a method for automatic recognition of samples in hip-hop music. A sample is defined as a short extraction from a source audio corpus that may have been embedded into another audio mixture. A series of non-negative matrix factorizations are applied to log-frequency spectrograms of hip-hop music and the source material from a master corpus. The factorizations result in matrices of base spectra and amplitude envelopes for the original and mixed audio. Each window of the mixed audio is compared to the original audio by examining the extracted amplitude envelopes. Several image-similarity metrics are employed to determine how closely the sampled and mixed amplitude envelopes match. Preliminary testing indicates that, as distinct from existing audio fingerprinting algorithms, the algorithm we describe is able to confirm instances of sampling in a hip-hop music mixture that the untrained listener is frequently unable to detect.

Original languageEnglish (US)
Title of host publication134th Audio Engineering Society Convention 2013
Pages852-860
Number of pages9
StatePublished - 2013
Event134th Audio Engineering Society Convention 2013 - Rome, Italy
Duration: May 4 2013May 7 2013

Other

Other134th Audio Engineering Society Convention 2013
CountryItaly
CityRome
Period5/4/135/7/13

Fingerprint

music
factorization
envelopes
matrices
spectrograms
sampling

ASJC Scopus subject areas

  • Acoustics and Ultrasonics

Cite this

Whitney, J. L., & Leider, C. (2013). Automatic sample recognition in hip-hop music based on non-negative matrix factorization. In 134th Audio Engineering Society Convention 2013 (pp. 852-860)

Automatic sample recognition in hip-hop music based on non-negative matrix factorization. / Whitney, Jordan L.; Leider, Colby.

134th Audio Engineering Society Convention 2013. 2013. p. 852-860.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Whitney, JL & Leider, C 2013, Automatic sample recognition in hip-hop music based on non-negative matrix factorization. in 134th Audio Engineering Society Convention 2013. pp. 852-860, 134th Audio Engineering Society Convention 2013, Rome, Italy, 5/4/13.
Whitney JL, Leider C. Automatic sample recognition in hip-hop music based on non-negative matrix factorization. In 134th Audio Engineering Society Convention 2013. 2013. p. 852-860
Whitney, Jordan L. ; Leider, Colby. / Automatic sample recognition in hip-hop music based on non-negative matrix factorization. 134th Audio Engineering Society Convention 2013. 2013. pp. 852-860
@inproceedings{0d9f3f90b92d4739a8029ea864a4a33a,
title = "Automatic sample recognition in hip-hop music based on non-negative matrix factorization",
abstract = "We present a method for automatic recognition of samples in hip-hop music. A sample is defined as a short extraction from a source audio corpus that may have been embedded into another audio mixture. A series of non-negative matrix factorizations are applied to log-frequency spectrograms of hip-hop music and the source material from a master corpus. The factorizations result in matrices of base spectra and amplitude envelopes for the original and mixed audio. Each window of the mixed audio is compared to the original audio by examining the extracted amplitude envelopes. Several image-similarity metrics are employed to determine how closely the sampled and mixed amplitude envelopes match. Preliminary testing indicates that, as distinct from existing audio fingerprinting algorithms, the algorithm we describe is able to confirm instances of sampling in a hip-hop music mixture that the untrained listener is frequently unable to detect.",
author = "Whitney, {Jordan L.} and Colby Leider",
year = "2013",
language = "English (US)",
isbn = "9781627485715",
pages = "852--860",
booktitle = "134th Audio Engineering Society Convention 2013",

}

TY - GEN

T1 - Automatic sample recognition in hip-hop music based on non-negative matrix factorization

AU - Whitney, Jordan L.

AU - Leider, Colby

PY - 2013

Y1 - 2013

N2 - We present a method for automatic recognition of samples in hip-hop music. A sample is defined as a short extraction from a source audio corpus that may have been embedded into another audio mixture. A series of non-negative matrix factorizations are applied to log-frequency spectrograms of hip-hop music and the source material from a master corpus. The factorizations result in matrices of base spectra and amplitude envelopes for the original and mixed audio. Each window of the mixed audio is compared to the original audio by examining the extracted amplitude envelopes. Several image-similarity metrics are employed to determine how closely the sampled and mixed amplitude envelopes match. Preliminary testing indicates that, as distinct from existing audio fingerprinting algorithms, the algorithm we describe is able to confirm instances of sampling in a hip-hop music mixture that the untrained listener is frequently unable to detect.

AB - We present a method for automatic recognition of samples in hip-hop music. A sample is defined as a short extraction from a source audio corpus that may have been embedded into another audio mixture. A series of non-negative matrix factorizations are applied to log-frequency spectrograms of hip-hop music and the source material from a master corpus. The factorizations result in matrices of base spectra and amplitude envelopes for the original and mixed audio. Each window of the mixed audio is compared to the original audio by examining the extracted amplitude envelopes. Several image-similarity metrics are employed to determine how closely the sampled and mixed amplitude envelopes match. Preliminary testing indicates that, as distinct from existing audio fingerprinting algorithms, the algorithm we describe is able to confirm instances of sampling in a hip-hop music mixture that the untrained listener is frequently unable to detect.

UR - http://www.scopus.com/inward/record.url?scp=84883408900&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883408900&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84883408900

SN - 9781627485715

SP - 852

EP - 860

BT - 134th Audio Engineering Society Convention 2013

ER -