Music segmentation and summarization based on self-similarity matrix

Sanghoon Jun, Een Jun Hwang

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

2 Citations (Scopus)

Abstract

In this paper, we propose a new method for segmenting and summarizing music based on its structure analysis. To do that, we first extract timbre feature from acoustic music signal and construct a self-similarity matrix that shows similarities among the features within music clip. We then determine candidate boundaries for music segmentation by tracking standard deviation in the matrix. Similar segments such as repetition in music clip are clustered and merged. In this way, each music clip can be represented by a sequence of states where each state represents a music segment with similar feature. We assume that the longest segment of a music clip represents the music and hence use it as a summary of the music clip. We show the performance of our proposed method through experiments.

Original languageEnglish
Title of host publicationProceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013
DOIs
Publication statusPublished - 2013 Apr 10
Event7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013 - Kota Kinabalu, Malaysia
Duration: 2013 Jan 172013 Jan 19

Other

Other7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013
CountryMalaysia
CityKota Kinabalu
Period13/1/1713/1/19

Fingerprint

Acoustics
Experiments

Keywords

  • Music retrieval
  • Music segmentation
  • Music summarization
  • Selfsimilarity matrix

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems

Cite this

Jun, S., & Hwang, E. J. (2013). Music segmentation and summarization based on self-similarity matrix. In Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013 [82] https://doi.org/10.1145/2448556.2448638

Music segmentation and summarization based on self-similarity matrix. / Jun, Sanghoon; Hwang, Een Jun.

Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013. 2013. 82.

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

Jun, S & Hwang, EJ 2013, Music segmentation and summarization based on self-similarity matrix. in Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013., 82, 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013, Kota Kinabalu, Malaysia, 13/1/17. https://doi.org/10.1145/2448556.2448638
Jun S, Hwang EJ. Music segmentation and summarization based on self-similarity matrix. In Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013. 2013. 82 https://doi.org/10.1145/2448556.2448638
Jun, Sanghoon ; Hwang, Een Jun. / Music segmentation and summarization based on self-similarity matrix. Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013. 2013.
@inproceedings{5348dedfaa5a4e7e88e09229546e3614,
title = "Music segmentation and summarization based on self-similarity matrix",
abstract = "In this paper, we propose a new method for segmenting and summarizing music based on its structure analysis. To do that, we first extract timbre feature from acoustic music signal and construct a self-similarity matrix that shows similarities among the features within music clip. We then determine candidate boundaries for music segmentation by tracking standard deviation in the matrix. Similar segments such as repetition in music clip are clustered and merged. In this way, each music clip can be represented by a sequence of states where each state represents a music segment with similar feature. We assume that the longest segment of a music clip represents the music and hence use it as a summary of the music clip. We show the performance of our proposed method through experiments.",
keywords = "Music retrieval, Music segmentation, Music summarization, Selfsimilarity matrix",
author = "Sanghoon Jun and Hwang, {Een Jun}",
year = "2013",
month = "4",
day = "10",
doi = "10.1145/2448556.2448638",
language = "English",
isbn = "9781450319584",
booktitle = "Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013",

}

TY - GEN

T1 - Music segmentation and summarization based on self-similarity matrix

AU - Jun, Sanghoon

AU - Hwang, Een Jun

PY - 2013/4/10

Y1 - 2013/4/10

N2 - In this paper, we propose a new method for segmenting and summarizing music based on its structure analysis. To do that, we first extract timbre feature from acoustic music signal and construct a self-similarity matrix that shows similarities among the features within music clip. We then determine candidate boundaries for music segmentation by tracking standard deviation in the matrix. Similar segments such as repetition in music clip are clustered and merged. In this way, each music clip can be represented by a sequence of states where each state represents a music segment with similar feature. We assume that the longest segment of a music clip represents the music and hence use it as a summary of the music clip. We show the performance of our proposed method through experiments.

AB - In this paper, we propose a new method for segmenting and summarizing music based on its structure analysis. To do that, we first extract timbre feature from acoustic music signal and construct a self-similarity matrix that shows similarities among the features within music clip. We then determine candidate boundaries for music segmentation by tracking standard deviation in the matrix. Similar segments such as repetition in music clip are clustered and merged. In this way, each music clip can be represented by a sequence of states where each state represents a music segment with similar feature. We assume that the longest segment of a music clip represents the music and hence use it as a summary of the music clip. We show the performance of our proposed method through experiments.

KW - Music retrieval

KW - Music segmentation

KW - Music summarization

KW - Selfsimilarity matrix

UR - http://www.scopus.com/inward/record.url?scp=84875835479&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875835479&partnerID=8YFLogxK

U2 - 10.1145/2448556.2448638

DO - 10.1145/2448556.2448638

M3 - Conference contribution

AN - SCOPUS:84875835479

SN - 9781450319584

BT - Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013

ER -