Name
Abstract | ||
|---|---|---|
Melody is an important property for the perceptual description of Western musical pieces. A lot of applications rely on the evaluation of similarity between two melodies. While several existing techniques assume a monophonic context or extract a monophonic melody from polyphonic pieces, in this paper, we propose to consider the whole polyphonic context to evaluate the similarity without reducing to a monophonic melody. We thus propose a new model and a corresponding methodology that takes into account all the notes, even if they sound at the same time or if they overlap. Our model relies on a quotiented sequence representation of music. A quotiented sequence is a sequence graph defined with an additional equivalent relation on its vertices and such that the quotient graph is also a sequence graph. The core of the comparison method is based on an adaptation of edit-distance metrics, regularly applied in bio-informatic context. This algorithm is currently being used to evaluate the similarity between a monophonic or polyphonic query and a database of polyphonic musical pieces. First experiments show that the adaptation to polyphony does not degrade the quality of the algorithm with monophonic musical pieces. Furthermore, the results of experiments with polyphonic pieces are promising, even if they show some limitations. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1007/978-3-642-12439-6_24 | CMMR/ICAD |
Keywords | Field | DocType |
monophonic context,sequence graph,quotiented sequence representation,quotiented sequence,polyphonic piece,polyphonic query,monophonic musical piece,monophonic melody,symbolic music retrieval,polyphonic musical piece,polyphonic alignment algorithm,whole polyphonic context,equivalence relation,edit distance | Melody,Graph,Music information retrieval,Vertex (geometry),Computer science,Algorithm,Speech recognition,Artificial intelligence,Natural language processing,Polyphony,Quotient graph | Conference |
Volume | ISSN | ISBN |
5954 | 0302-9743 | 3-642-12438-0 |
Citations | PageRank | References |
2 | 0.43 | 15 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Julien Allali | 1 | 64 | 8.16 |
| Pascal Ferraro | 2 | 77 | 11.54 |
| Pierre Hanna | 3 | 110 | 20.53 |
| Matthias Robine | 4 | 74 | 13.06 |