Melody-based Knowledge Discovery in Musical Pieces

Agnieszka Jastrzębska , Rybnik Mariusz

Abstract

The paper is focused on automated knowledge discovery in musical pieces, based on transformations of digital musical notation. Usually a single musical piece is analyzed, to discover the structure as well as traits of separate voices. Melody and rhythm is processed with the use of three proposed operators, that serve as meta-data. In this work we focus on melody, so the processed data is labeled using fuzzy labels, created for detecting various voice characteristics. A comparative analysis of two musical pieces may be performed as well, that compares them in terms of various rhythmic or melodic traits (as a whole or with voice separation).
Author Agnieszka Jastrzębska ZSMPW
Agnieszka Jastrzębska,,
- Department of Structural Methods for Knowledge Processing
, Rybnik Mariusz
Rybnik Mariusz,,
-
Pages1800061-1800064
Publication size in sheets90003.05
Book Simos Theodore, Tsitouras Charalambos (eds.): Proceedings of The International Conference on Numerical Analysis and Applied Mathematics 2015 (ICNAAM-2015), AIP Conference Proceedings, vol. 1738, 2016, AMER INST PHYSICS, ISBN 978-0-7354-1392-4
Keywords in EnglishMusic Information Retrieval; searching; knowledge discovery; knowledge understanding
Abstract in PolishW pracy omówiono zagadnienie automatycznego przetwarzania notacji muzycznej. Celem badań przedstawionych w pracy było stworzenie metod „głębokiej” analizy utworu, w celu wykrycia jego struktury. Melodia oraz rytm zostały opisane za pomocą trzech zaproponowanych operatorów. W tej pracy w szczególności rozważana jest melodia, a w celu lepszego jej opisu zastosowano teorię zbiorów rozmytych.
DOIDOI:10.1063/1.4951953
URL http://aip.scitation.org/doi/abs/10.1063/1.4951953
Languageen angielski
Score (nominal)15
ScoreMinisterial score = 15.0, 26-06-2017, BookChapterSeriesAndMatConf
Ministerial score (2013-2016) = 15.0, 26-06-2017, BookChapterSeriesAndMatConf
Citation count*0
Cite
Share Share



* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back