Improvements of Audio-based Music Similarity and Genre Classification

Elias Pampalk; Arthur Flexer; Gerhard Widmer

Improvements of Audio-based Music Similarity and Genre Classification

Elias Pampalk, Arthur Flexer, Gerhard Widmer

Institute of Computational Perception

Research output: Chapter in Book/Report/Conference proceeding › Conference proceedings › peer-review

Abstract

Audio-based music similarity measures can be used to automatically generate playlists or recommendations. In this paper the similarity measure that won the ISMIR04 genre classification contest is reviewed. In addition, further improvements are presented. In particular, two new descriptors are presented and combined with two previously published similarity measures. The performance is evaluated in a series of experiments on four music collections. The evaluations are based on genre classification, assuming that very similar tracks belong to the same genre. On two collections the improvements lead to a substantial performance increase.

Original language	English
Title of host publication	Proceedings of the 6th International Conference on Music Information Retrieval (ISMIR 2005), London, UK
Number of pages	8
Publication status	Published - 2005

Fields of science

102 Computer Sciences
102001 Artificial intelligence
102003 Image processing
102015 Information systems
202002 Audiovisual media

Cite this

@inproceedings{7dc2d8ae0cf546d1bf4226605d36a653,

title = "Improvements of Audio-based Music Similarity and Genre Classification",

abstract = "Audio-based music similarity measures can be used to automatically generate playlists or recommendations. In this paper the similarity measure that won the ISMIR04 genre classification contest is reviewed. In addition, further improvements are presented. In particular, two new descriptors are presented and combined with two previously published similarity measures. The performance is evaluated in a series of experiments on four music collections. The evaluations are based on genre classification, assuming that very similar tracks belong to the same genre. On two collections the improvements lead to a substantial performance increase.",

author = "Elias Pampalk and Arthur Flexer and Gerhard Widmer",

year = "2005",

language = "English",

booktitle = "Proceedings of the 6th International Conference on Music Information Retrieval (ISMIR 2005), London, UK",

}

TY - GEN

T1 - Improvements of Audio-based Music Similarity and Genre Classification

AU - Pampalk, Elias

AU - Flexer, Arthur

AU - Widmer, Gerhard

PY - 2005

Y1 - 2005

N2 - Audio-based music similarity measures can be used to automatically generate playlists or recommendations. In this paper the similarity measure that won the ISMIR04 genre classification contest is reviewed. In addition, further improvements are presented. In particular, two new descriptors are presented and combined with two previously published similarity measures. The performance is evaluated in a series of experiments on four music collections. The evaluations are based on genre classification, assuming that very similar tracks belong to the same genre. On two collections the improvements lead to a substantial performance increase.

AB - Audio-based music similarity measures can be used to automatically generate playlists or recommendations. In this paper the similarity measure that won the ISMIR04 genre classification contest is reviewed. In addition, further improvements are presented. In particular, two new descriptors are presented and combined with two previously published similarity measures. The performance is evaluated in a series of experiments on four music collections. The evaluations are based on genre classification, assuming that very similar tracks belong to the same genre. On two collections the improvements lead to a substantial performance increase.

M3 - Conference proceedings

BT - Proceedings of the 6th International Conference on Music Information Retrieval (ISMIR 2005), London, UK

ER -