LyricSRadar: A lyrics retrieval system based on latent topics of lyrics

Shoto Sasaki, Kazuyoshi Yoshii, Tomoyasu Nakano, Masataka Goto, Shigeo Morishima

Research output: Contribution to conferencePaper

10 Citations (Scopus)

Abstract

This paper presents a lyrics retrieval system called LyricsRadar that enables users to interactively browse song lyrics by visualizing their topics. Since conventional lyrics retrieval systems are based on simple word search, those systems often fail to reflect user’s intention behind a query when a word given as a query can be used in different contexts. For example, the wordtearscan appear not only in sad songs (e.g., feel heartrending), but also in happy songs (e.g., weep for joy). To overcome this limitation, we propose to automatically analyze and visualize topics of lyrics by using a well-known text analysis method called latent Dirichlet allocation (LDA). This enables LyricsRadar to offer two types of topic visualization. One is the topic radar chart that visualizes the relative weights of five latent topics of each song on a pentagon-shaped chart. The other is radar-like arrangement of all songs in a two-dimensional space in which song lyrics having similar topics are arranged close to each other. The subjective experiments using 6,902 Japanese popular songs showed that our system can appropriately navigate users to lyrics of interests.

Original languageEnglish
Pages585-590
Number of pages6
Publication statusPublished - 2014 Jan 1
Event15th International Society for Music Information Retrieval Conference, ISMIR 2014 - Taipei, Taiwan, Province of China
Duration: 2014 Oct 272014 Oct 31

Conference

Conference15th International Society for Music Information Retrieval Conference, ISMIR 2014
CountryTaiwan, Province of China
CityTaipei
Period14/10/2714/10/31

Fingerprint

Radar
Visualization
Experiments
Lyrics
Song
Song Lyrics
Charts

ASJC Scopus subject areas

  • Music
  • Information Systems

Cite this

Sasaki, S., Yoshii, K., Nakano, T., Goto, M., & Morishima, S. (2014). LyricSRadar: A lyrics retrieval system based on latent topics of lyrics. 585-590. Paper presented at 15th International Society for Music Information Retrieval Conference, ISMIR 2014, Taipei, Taiwan, Province of China.

LyricSRadar : A lyrics retrieval system based on latent topics of lyrics. / Sasaki, Shoto; Yoshii, Kazuyoshi; Nakano, Tomoyasu; Goto, Masataka; Morishima, Shigeo.

2014. 585-590 Paper presented at 15th International Society for Music Information Retrieval Conference, ISMIR 2014, Taipei, Taiwan, Province of China.

Research output: Contribution to conferencePaper

Sasaki, S, Yoshii, K, Nakano, T, Goto, M & Morishima, S 2014, 'LyricSRadar: A lyrics retrieval system based on latent topics of lyrics' Paper presented at 15th International Society for Music Information Retrieval Conference, ISMIR 2014, Taipei, Taiwan, Province of China, 14/10/27 - 14/10/31, pp. 585-590.
Sasaki S, Yoshii K, Nakano T, Goto M, Morishima S. LyricSRadar: A lyrics retrieval system based on latent topics of lyrics. 2014. Paper presented at 15th International Society for Music Information Retrieval Conference, ISMIR 2014, Taipei, Taiwan, Province of China.
Sasaki, Shoto ; Yoshii, Kazuyoshi ; Nakano, Tomoyasu ; Goto, Masataka ; Morishima, Shigeo. / LyricSRadar : A lyrics retrieval system based on latent topics of lyrics. Paper presented at 15th International Society for Music Information Retrieval Conference, ISMIR 2014, Taipei, Taiwan, Province of China.6 p.
@conference{858e638ffd19416097db7b6aac050b01,
title = "LyricSRadar: A lyrics retrieval system based on latent topics of lyrics",
abstract = "This paper presents a lyrics retrieval system called LyricsRadar that enables users to interactively browse song lyrics by visualizing their topics. Since conventional lyrics retrieval systems are based on simple word search, those systems often fail to reflect user’s intention behind a query when a word given as a query can be used in different contexts. For example, the wordtearscan appear not only in sad songs (e.g., feel heartrending), but also in happy songs (e.g., weep for joy). To overcome this limitation, we propose to automatically analyze and visualize topics of lyrics by using a well-known text analysis method called latent Dirichlet allocation (LDA). This enables LyricsRadar to offer two types of topic visualization. One is the topic radar chart that visualizes the relative weights of five latent topics of each song on a pentagon-shaped chart. The other is radar-like arrangement of all songs in a two-dimensional space in which song lyrics having similar topics are arranged close to each other. The subjective experiments using 6,902 Japanese popular songs showed that our system can appropriately navigate users to lyrics of interests.",
author = "Shoto Sasaki and Kazuyoshi Yoshii and Tomoyasu Nakano and Masataka Goto and Shigeo Morishima",
year = "2014",
month = "1",
day = "1",
language = "English",
pages = "585--590",
note = "15th International Society for Music Information Retrieval Conference, ISMIR 2014 ; Conference date: 27-10-2014 Through 31-10-2014",

}

TY - CONF

T1 - LyricSRadar

T2 - A lyrics retrieval system based on latent topics of lyrics

AU - Sasaki, Shoto

AU - Yoshii, Kazuyoshi

AU - Nakano, Tomoyasu

AU - Goto, Masataka

AU - Morishima, Shigeo

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper presents a lyrics retrieval system called LyricsRadar that enables users to interactively browse song lyrics by visualizing their topics. Since conventional lyrics retrieval systems are based on simple word search, those systems often fail to reflect user’s intention behind a query when a word given as a query can be used in different contexts. For example, the wordtearscan appear not only in sad songs (e.g., feel heartrending), but also in happy songs (e.g., weep for joy). To overcome this limitation, we propose to automatically analyze and visualize topics of lyrics by using a well-known text analysis method called latent Dirichlet allocation (LDA). This enables LyricsRadar to offer two types of topic visualization. One is the topic radar chart that visualizes the relative weights of five latent topics of each song on a pentagon-shaped chart. The other is radar-like arrangement of all songs in a two-dimensional space in which song lyrics having similar topics are arranged close to each other. The subjective experiments using 6,902 Japanese popular songs showed that our system can appropriately navigate users to lyrics of interests.

AB - This paper presents a lyrics retrieval system called LyricsRadar that enables users to interactively browse song lyrics by visualizing their topics. Since conventional lyrics retrieval systems are based on simple word search, those systems often fail to reflect user’s intention behind a query when a word given as a query can be used in different contexts. For example, the wordtearscan appear not only in sad songs (e.g., feel heartrending), but also in happy songs (e.g., weep for joy). To overcome this limitation, we propose to automatically analyze and visualize topics of lyrics by using a well-known text analysis method called latent Dirichlet allocation (LDA). This enables LyricsRadar to offer two types of topic visualization. One is the topic radar chart that visualizes the relative weights of five latent topics of each song on a pentagon-shaped chart. The other is radar-like arrangement of all songs in a two-dimensional space in which song lyrics having similar topics are arranged close to each other. The subjective experiments using 6,902 Japanese popular songs showed that our system can appropriately navigate users to lyrics of interests.

UR - http://www.scopus.com/inward/record.url?scp=84951980403&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951980403&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84951980403

SP - 585

EP - 590

ER -