Maximum entropy models capture melodic styles

Abstract We introduce a Maximum Entropy model able to capture the statistics of melodies in music. The model can be used to generate new melodies that emulate the style of a given musical corpus. Instead of using the n–body interactions of (n−1)–order Markov models, traditionally used in automatic m...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jason Sakellariou, Francesca Tria, Vittorio Loreto, Francois Pachet
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2017
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/33531b39a72f401f90ac055edba3da06
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:33531b39a72f401f90ac055edba3da06
record_format	dspace
spelling	oai:doaj.org-article:33531b39a72f401f90ac055edba3da062021-12-02T11:40:22ZMaximum entropy models capture melodic styles10.1038/s41598-017-08028-42045-2322https://doaj.org/article/33531b39a72f401f90ac055edba3da062017-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-017-08028-4https://doaj.org/toc/2045-2322Abstract We introduce a Maximum Entropy model able to capture the statistics of melodies in music. The model can be used to generate new melodies that emulate the style of a given musical corpus. Instead of using the n–body interactions of (n−1)–order Markov models, traditionally used in automatic music generation, we use a k-nearest neighbour model with pairwise interactions only. In that way, we keep the number of parameters low and avoid over-fitting problems typical of Markov models. We show that long-range musical phrases don’t need to be explicitly enforced using high-order Markov interactions, but can instead emerge from multiple, competing, pairwise interactions. We validate our Maximum Entropy model by contrasting how much the generated sequences capture the style of the original corpus without plagiarizing it. To this end we use a data-compression approach to discriminate the levels of borrowing and innovation featured by the artificial sequences. Our modelling scheme outperforms both fixed-order and variable-order Markov models. This shows that, despite being based only on pairwise interactions, our scheme opens the possibility to generate musically sensible alterations of the original phrases, providing a way to generate innovation.Jason SakellariouFrancesca TriaVittorio LoretoFrancois PachetNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 7, Iss 1, Pp 1-9 (2017)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Jason Sakellariou Francesca Tria Vittorio Loreto Francois Pachet Maximum entropy models capture melodic styles
description	Abstract We introduce a Maximum Entropy model able to capture the statistics of melodies in music. The model can be used to generate new melodies that emulate the style of a given musical corpus. Instead of using the n–body interactions of (n−1)–order Markov models, traditionally used in automatic music generation, we use a k-nearest neighbour model with pairwise interactions only. In that way, we keep the number of parameters low and avoid over-fitting problems typical of Markov models. We show that long-range musical phrases don’t need to be explicitly enforced using high-order Markov interactions, but can instead emerge from multiple, competing, pairwise interactions. We validate our Maximum Entropy model by contrasting how much the generated sequences capture the style of the original corpus without plagiarizing it. To this end we use a data-compression approach to discriminate the levels of borrowing and innovation featured by the artificial sequences. Our modelling scheme outperforms both fixed-order and variable-order Markov models. This shows that, despite being based only on pairwise interactions, our scheme opens the possibility to generate musically sensible alterations of the original phrases, providing a way to generate innovation.
format	article
author	Jason Sakellariou Francesca Tria Vittorio Loreto Francois Pachet
author_facet	Jason Sakellariou Francesca Tria Vittorio Loreto Francois Pachet
author_sort	Jason Sakellariou
title	Maximum entropy models capture melodic styles
title_short	Maximum entropy models capture melodic styles
title_full	Maximum entropy models capture melodic styles
title_fullStr	Maximum entropy models capture melodic styles
title_full_unstemmed	Maximum entropy models capture melodic styles
title_sort	maximum entropy models capture melodic styles
publisher	Nature Portfolio
publishDate	2017
url	https://doaj.org/article/33531b39a72f401f90ac055edba3da06
work_keys_str_mv	AT jasonsakellariou maximumentropymodelscapturemelodicstyles AT francescatria maximumentropymodelscapturemelodicstyles AT vittorioloreto maximumentropymodelscapturemelodicstyles AT francoispachet maximumentropymodelscapturemelodicstyles
_version_	1718395634267979776

Maximum entropy models capture melodic styles

Ejemplares similares