Creating ground truth for audio key finding

When the title key may not be the key

Ching-Hua Chuan, Elaine Chew

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

3 Citations (Scopus)

Abstract

In this paper, we present an effective and efficient way to create an accurately labeled dataset to advance audio key finding research. The MIREX audio key finding contest has been held twice using classical compositions for which the key is designated in the title. The problem with this accepted practice is that the title key may not be the perceived key in the audio excerpt. To reduce manual annotation, which is costly, we use a confusion index generated by existing audio key finding algorithms to determine if an audio excerpt requires manual annotation. We collected 3224 excerpts and identified 727 excerpts requiring manual annotation. We evaluate the algorithms' performance on these challenging cases using the title keys, and the re-labeled keys. The musicians who aurally identify the key also provide comments on the reasons for their choice. The relabeling process reveals the mismatch between title and perceived keys to be caused by tuning practices (in 471 of the 727 excerpts, 64.79%), and other factors (188 excerpts, 25.86%) including key modulation and intonation choices. The remaining 68 challenging cases provide useful information for algorithm design.

Original languageEnglish (US)
Title of host publicationProceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012
Pages247-252
Number of pages6
StatePublished - Dec 1 2012
Externally publishedYes
Event13th International Society for Music Information Retrieval Conference, ISMIR 2012 - Porto, Portugal
Duration: Oct 8 2012Oct 12 2012

Other

Other13th International Society for Music Information Retrieval Conference, ISMIR 2012
CountryPortugal
CityPorto
Period10/8/1210/12/12

Fingerprint

Tuning
Modulation
Chemical analysis
Annotation
Mismatch
Confusion
Intonation
Contests
Musicians

ASJC Scopus subject areas

  • Music
  • Information Systems

Cite this

Chuan, C-H., & Chew, E. (2012). Creating ground truth for audio key finding: When the title key may not be the key. In Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012 (pp. 247-252)

Creating ground truth for audio key finding : When the title key may not be the key. / Chuan, Ching-Hua; Chew, Elaine.

Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012. 2012. p. 247-252.

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

Chuan, C-H & Chew, E 2012, Creating ground truth for audio key finding: When the title key may not be the key. in Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012. pp. 247-252, 13th International Society for Music Information Retrieval Conference, ISMIR 2012, Porto, Portugal, 10/8/12.
Chuan C-H, Chew E. Creating ground truth for audio key finding: When the title key may not be the key. In Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012. 2012. p. 247-252
Chuan, Ching-Hua ; Chew, Elaine. / Creating ground truth for audio key finding : When the title key may not be the key. Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012. 2012. pp. 247-252
@inproceedings{0ddee202405542bcafdb76dde517ef8e,
title = "Creating ground truth for audio key finding: When the title key may not be the key",
abstract = "In this paper, we present an effective and efficient way to create an accurately labeled dataset to advance audio key finding research. The MIREX audio key finding contest has been held twice using classical compositions for which the key is designated in the title. The problem with this accepted practice is that the title key may not be the perceived key in the audio excerpt. To reduce manual annotation, which is costly, we use a confusion index generated by existing audio key finding algorithms to determine if an audio excerpt requires manual annotation. We collected 3224 excerpts and identified 727 excerpts requiring manual annotation. We evaluate the algorithms' performance on these challenging cases using the title keys, and the re-labeled keys. The musicians who aurally identify the key also provide comments on the reasons for their choice. The relabeling process reveals the mismatch between title and perceived keys to be caused by tuning practices (in 471 of the 727 excerpts, 64.79{\%}), and other factors (188 excerpts, 25.86{\%}) including key modulation and intonation choices. The remaining 68 challenging cases provide useful information for algorithm design.",
author = "Ching-Hua Chuan and Elaine Chew",
year = "2012",
month = "12",
day = "1",
language = "English (US)",
isbn = "9789727521449",
pages = "247--252",
booktitle = "Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012",

}

TY - GEN

T1 - Creating ground truth for audio key finding

T2 - When the title key may not be the key

AU - Chuan, Ching-Hua

AU - Chew, Elaine

PY - 2012/12/1

Y1 - 2012/12/1

N2 - In this paper, we present an effective and efficient way to create an accurately labeled dataset to advance audio key finding research. The MIREX audio key finding contest has been held twice using classical compositions for which the key is designated in the title. The problem with this accepted practice is that the title key may not be the perceived key in the audio excerpt. To reduce manual annotation, which is costly, we use a confusion index generated by existing audio key finding algorithms to determine if an audio excerpt requires manual annotation. We collected 3224 excerpts and identified 727 excerpts requiring manual annotation. We evaluate the algorithms' performance on these challenging cases using the title keys, and the re-labeled keys. The musicians who aurally identify the key also provide comments on the reasons for their choice. The relabeling process reveals the mismatch between title and perceived keys to be caused by tuning practices (in 471 of the 727 excerpts, 64.79%), and other factors (188 excerpts, 25.86%) including key modulation and intonation choices. The remaining 68 challenging cases provide useful information for algorithm design.

AB - In this paper, we present an effective and efficient way to create an accurately labeled dataset to advance audio key finding research. The MIREX audio key finding contest has been held twice using classical compositions for which the key is designated in the title. The problem with this accepted practice is that the title key may not be the perceived key in the audio excerpt. To reduce manual annotation, which is costly, we use a confusion index generated by existing audio key finding algorithms to determine if an audio excerpt requires manual annotation. We collected 3224 excerpts and identified 727 excerpts requiring manual annotation. We evaluate the algorithms' performance on these challenging cases using the title keys, and the re-labeled keys. The musicians who aurally identify the key also provide comments on the reasons for their choice. The relabeling process reveals the mismatch between title and perceived keys to be caused by tuning practices (in 471 of the 727 excerpts, 64.79%), and other factors (188 excerpts, 25.86%) including key modulation and intonation choices. The remaining 68 challenging cases provide useful information for algorithm design.

UR - http://www.scopus.com/inward/record.url?scp=84873455583&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873455583&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9789727521449

SP - 247

EP - 252

BT - Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012

ER -