Exploring melodic motif to support an affect-based music compositional intelligence

Roberto Legaspi, Akinobu Ueda, Rafael Cabredo, Takayuki Nishikawa, Kenichi Fukui, Koichi Moriyama, Satoshi Kurihara, Masayuki Numao

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

Abstract

Although the design of our constructive adaptive user interface (CAUI) for an affect-based music compositional artificial intelligence has been modified on several fronts since the time it was introduced, what has become a persisting limitation of our research is the extent by which it should efficiently cover music theory effectively. This paper reports our initial investigation on the possible significant contribution of melodic motif in creating compositions that are more fluent and cohesive. From an initial collection of 10 melodic motifs from different musical pieces, we provided heuristic-based renditions to these melodic motifs, four for each one, and obtained a total of 50 melodic motifs. We asked 10 subjects to provide self-annotations of the affective flavor of these motifs. We then represented these motifs as first-order logic predicates and employed inductive logic programming for the CAUI to learn relations of user affect perceptions and music features. To obtain new compositions, we first used a genetic algorithm with a fitness function that is based on the induced relations for the CAUI to generate chordal tone variants. We then used probabilistic modifications for the CAUI to alter these chordal tones to become non-harmonic tones. The CAUI composed 60 new user-specific affect-based musical pieces for each subject. Our results indicate that the compositions differ significantly for only one pair of affect type when the subject evaluations of the CAUI compositions were compared using paired t-test. However, when we compared the subject evaluations of the quality of the melodies and of the musical pieces from when melodic motif variants were not considered, the improvement is significant with t-values of 5.86 and 6.33, respectively, for a significance level of 0.01.

Original languageEnglish
Title of host publicationProceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011
Pages219-225
Number of pages7
DOIs
Publication statusPublished - 2011 Nov 21
Externally publishedYes
Event2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011 - Hanoi, Viet Nam
Duration: 2011 Oct 142011 Oct 17

Other

Other2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011
CountryViet Nam
CityHanoi
Period11/10/1411/10/17

Fingerprint

User interfaces
Chemical analysis
Inductive logic programming (ILP)
Flavors
Artificial intelligence
Genetic algorithms

Keywords

  • emotion recognition
  • human-computer interaction
  • music information retrieval

ASJC Scopus subject areas

  • Information Systems
  • Software

Cite this

Legaspi, R., Ueda, A., Cabredo, R., Nishikawa, T., Fukui, K., Moriyama, K., ... Numao, M. (2011). Exploring melodic motif to support an affect-based music compositional intelligence. In Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011 (pp. 219-225). [6063470] https://doi.org/10.1109/KSE.2011.42

Exploring melodic motif to support an affect-based music compositional intelligence. / Legaspi, Roberto; Ueda, Akinobu; Cabredo, Rafael; Nishikawa, Takayuki; Fukui, Kenichi; Moriyama, Koichi; Kurihara, Satoshi; Numao, Masayuki.

Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011. 2011. p. 219-225 6063470.

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

Legaspi, R, Ueda, A, Cabredo, R, Nishikawa, T, Fukui, K, Moriyama, K, Kurihara, S & Numao, M 2011, Exploring melodic motif to support an affect-based music compositional intelligence. in Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011., 6063470, pp. 219-225, 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011, Hanoi, Viet Nam, 11/10/14. https://doi.org/10.1109/KSE.2011.42
Legaspi R, Ueda A, Cabredo R, Nishikawa T, Fukui K, Moriyama K et al. Exploring melodic motif to support an affect-based music compositional intelligence. In Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011. 2011. p. 219-225. 6063470 https://doi.org/10.1109/KSE.2011.42
Legaspi, Roberto ; Ueda, Akinobu ; Cabredo, Rafael ; Nishikawa, Takayuki ; Fukui, Kenichi ; Moriyama, Koichi ; Kurihara, Satoshi ; Numao, Masayuki. / Exploring melodic motif to support an affect-based music compositional intelligence. Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011. 2011. pp. 219-225
@inproceedings{b3cfad10237f4e90811999a08f0469ca,
title = "Exploring melodic motif to support an affect-based music compositional intelligence",
abstract = "Although the design of our constructive adaptive user interface (CAUI) for an affect-based music compositional artificial intelligence has been modified on several fronts since the time it was introduced, what has become a persisting limitation of our research is the extent by which it should efficiently cover music theory effectively. This paper reports our initial investigation on the possible significant contribution of melodic motif in creating compositions that are more fluent and cohesive. From an initial collection of 10 melodic motifs from different musical pieces, we provided heuristic-based renditions to these melodic motifs, four for each one, and obtained a total of 50 melodic motifs. We asked 10 subjects to provide self-annotations of the affective flavor of these motifs. We then represented these motifs as first-order logic predicates and employed inductive logic programming for the CAUI to learn relations of user affect perceptions and music features. To obtain new compositions, we first used a genetic algorithm with a fitness function that is based on the induced relations for the CAUI to generate chordal tone variants. We then used probabilistic modifications for the CAUI to alter these chordal tones to become non-harmonic tones. The CAUI composed 60 new user-specific affect-based musical pieces for each subject. Our results indicate that the compositions differ significantly for only one pair of affect type when the subject evaluations of the CAUI compositions were compared using paired t-test. However, when we compared the subject evaluations of the quality of the melodies and of the musical pieces from when melodic motif variants were not considered, the improvement is significant with t-values of 5.86 and 6.33, respectively, for a significance level of 0.01.",
keywords = "emotion recognition, human-computer interaction, music information retrieval",
author = "Roberto Legaspi and Akinobu Ueda and Rafael Cabredo and Takayuki Nishikawa and Kenichi Fukui and Koichi Moriyama and Satoshi Kurihara and Masayuki Numao",
year = "2011",
month = "11",
day = "21",
doi = "10.1109/KSE.2011.42",
language = "English",
isbn = "9780769545677",
pages = "219--225",
booktitle = "Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011",

}

TY - GEN

T1 - Exploring melodic motif to support an affect-based music compositional intelligence

AU - Legaspi, Roberto

AU - Ueda, Akinobu

AU - Cabredo, Rafael

AU - Nishikawa, Takayuki

AU - Fukui, Kenichi

AU - Moriyama, Koichi

AU - Kurihara, Satoshi

AU - Numao, Masayuki

PY - 2011/11/21

Y1 - 2011/11/21

N2 - Although the design of our constructive adaptive user interface (CAUI) for an affect-based music compositional artificial intelligence has been modified on several fronts since the time it was introduced, what has become a persisting limitation of our research is the extent by which it should efficiently cover music theory effectively. This paper reports our initial investigation on the possible significant contribution of melodic motif in creating compositions that are more fluent and cohesive. From an initial collection of 10 melodic motifs from different musical pieces, we provided heuristic-based renditions to these melodic motifs, four for each one, and obtained a total of 50 melodic motifs. We asked 10 subjects to provide self-annotations of the affective flavor of these motifs. We then represented these motifs as first-order logic predicates and employed inductive logic programming for the CAUI to learn relations of user affect perceptions and music features. To obtain new compositions, we first used a genetic algorithm with a fitness function that is based on the induced relations for the CAUI to generate chordal tone variants. We then used probabilistic modifications for the CAUI to alter these chordal tones to become non-harmonic tones. The CAUI composed 60 new user-specific affect-based musical pieces for each subject. Our results indicate that the compositions differ significantly for only one pair of affect type when the subject evaluations of the CAUI compositions were compared using paired t-test. However, when we compared the subject evaluations of the quality of the melodies and of the musical pieces from when melodic motif variants were not considered, the improvement is significant with t-values of 5.86 and 6.33, respectively, for a significance level of 0.01.

AB - Although the design of our constructive adaptive user interface (CAUI) for an affect-based music compositional artificial intelligence has been modified on several fronts since the time it was introduced, what has become a persisting limitation of our research is the extent by which it should efficiently cover music theory effectively. This paper reports our initial investigation on the possible significant contribution of melodic motif in creating compositions that are more fluent and cohesive. From an initial collection of 10 melodic motifs from different musical pieces, we provided heuristic-based renditions to these melodic motifs, four for each one, and obtained a total of 50 melodic motifs. We asked 10 subjects to provide self-annotations of the affective flavor of these motifs. We then represented these motifs as first-order logic predicates and employed inductive logic programming for the CAUI to learn relations of user affect perceptions and music features. To obtain new compositions, we first used a genetic algorithm with a fitness function that is based on the induced relations for the CAUI to generate chordal tone variants. We then used probabilistic modifications for the CAUI to alter these chordal tones to become non-harmonic tones. The CAUI composed 60 new user-specific affect-based musical pieces for each subject. Our results indicate that the compositions differ significantly for only one pair of affect type when the subject evaluations of the CAUI compositions were compared using paired t-test. However, when we compared the subject evaluations of the quality of the melodies and of the musical pieces from when melodic motif variants were not considered, the improvement is significant with t-values of 5.86 and 6.33, respectively, for a significance level of 0.01.

KW - emotion recognition

KW - human-computer interaction

KW - music information retrieval

UR - http://www.scopus.com/inward/record.url?scp=81255170022&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81255170022&partnerID=8YFLogxK

U2 - 10.1109/KSE.2011.42

DO - 10.1109/KSE.2011.42

M3 - Conference contribution

SN - 9780769545677

SP - 219

EP - 225

BT - Proceedings - 2011 3rd International Conference on Knowledge and Systems Engineering, KSE 2011

ER -