Local Transpositions in Alignment of Polyphonic Musical Sequences

Music retrieval systems based on melodic similarity consider sequences of notes. Adaptations of edit-based algorithms, mainly applied in bioinformatic applications, to the musical domain lead to promising results. However, new problems are raised when considering polyphonic music. Existing represent...

Full description

Saved in:
Bibliographic Details
Published inString Processing and Information Retrieval Vol. 4726; pp. 26 - 38
Main Authors Allali, Julien, Ferraro, Pascal, Hanna, Pierre, Iliopoulos, Costas
Format Book Chapter
LanguageEnglish
Published Germany Springer Berlin / Heidelberg 2007
Springer Berlin Heidelberg
SeriesLecture Notes in Computer Science
Subjects
Online AccessGet full text
ISBN9783540755296
3540755292
ISSN0302-9743
1611-3349
DOI10.1007/978-3-540-75530-2_3

Cover

More Information
Summary:Music retrieval systems based on melodic similarity consider sequences of notes. Adaptations of edit-based algorithms, mainly applied in bioinformatic applications, to the musical domain lead to promising results. However, new problems are raised when considering polyphonic music. Existing representations of notes do not allow retrieval systems to be transposition invariant. In this article, we propose a new dynamic programming algorithm that permits to take into account multiple local transpositions. Experiments with MIREX collections have been performed to evaluate the improvements induced by this algorithm. The results clearly show the contribution of this algorithm: it is confirmed as the most accurate solution for a music retrieval system based on alignment algorithm to be transposition invariant.
ISBN:9783540755296
3540755292
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-540-75530-2_3