Detecting solo phrases in music using spectral and pitch-related descriptors

In this paper we present an algorithm for segmenting musical audio data. Our aim is to identify solo instrument phrases in polyphonic music. We extract relevant features from the audio to be input into our algorithm. A large corpus of audio descriptors was tested for its ability to discriminate betw...

ver descrição completa

Detalhes bibliográficos
Autores: Fuhrmann, Ferdinand, Herrera Boyer, Perfecto, 1964-, Serra, Xavier
Tipo de documento: artigo
Estado:Versión aceptada para publicación
Data de publicação:2009
País:España
Recursos:Universitat Pompeu Fabra
Repositório:Repositorio Digital de la UPF
OAI Identifier:oai:repositori.upf.edu:10230/34624
Acesso em linha:http://hdl.handle.net/10230/34624
http://dx.doi.org/10.1080/09298210903406632
Access Level:Acceso aberto
Palavra-chave:So -- Informàtica
id ES_b41cba9eaf324df1128b6ee7e9f577a0
oai_identifier_str oai:repositori.upf.edu:10230/34624
network_acronym_str ES
network_name_str España
repository_id_str
spelling Detecting solo phrases in music using spectral and pitch-related descriptorsFuhrmann, FerdinandHerrera Boyer, Perfecto, 1964-Serra, XavierSo -- InformàticaIn this paper we present an algorithm for segmenting musical audio data. Our aim is to identify solo instrument phrases in polyphonic music. We extract relevant features from the audio to be input into our algorithm. A large corpus of audio descriptors was tested for its ability to discriminate between solo and non-solo sections, which resulted in a subset of five best features. We derived a two-stage algorithm that first creates a set of boundary candidates from local changes of these features and then classifies fixed-length segments according to the desired target classes. The output of the two stages is combined to derive the final segmentation and segment labels. Our system was trained and tested with excerpts from classical pieces and evaluated using full-length recordings, all taken from commercially available audio. We evaluated our algorithm by using precision and recall measurements for the boundary estimation and introduced new evaluation metrics from image processing for the final segmentation. Along with a resulting accuracy of 77%, we demonstrate that the selected features are discriminative for this specific task and achieve reasonable results for the segmentation problem.This research has been partially funded by the EU-IP project PHAROS.Taylor & Francis (Routledge)201820182009info:eu-repo/semantics/articleinfo:eu-repo/semantics/acceptedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/10230/34624http://dx.doi.org/10.1080/09298210903406632reponame:Repositorio Digital de la UPFinstname:Universitat Pompeu FabraInglésJournal of New Music Research. 2009;38(4):343-56© Taylor & Francis. This is an electronic version of an article published in Fuhrmann F, Herrera P, Serra X. Detecting solo phrases in music using spectral and pitch-related descriptors. J New Music Res. 2009;38(4):343-56. Journal of New Music Research is available online at: https://www.tandfonline.com/doi/abs/10.1080/09298210903406632.info:eu-repo/semantics/openAccessoai:repositori.upf.edu:10230/346242026-06-12T07:21:37Z
dc.title.none.fl_str_mv Detecting solo phrases in music using spectral and pitch-related descriptors
title Detecting solo phrases in music using spectral and pitch-related descriptors
spellingShingle Detecting solo phrases in music using spectral and pitch-related descriptors
Fuhrmann, Ferdinand
So -- Informàtica
title_short Detecting solo phrases in music using spectral and pitch-related descriptors
title_full Detecting solo phrases in music using spectral and pitch-related descriptors
title_fullStr Detecting solo phrases in music using spectral and pitch-related descriptors
title_full_unstemmed Detecting solo phrases in music using spectral and pitch-related descriptors
title_sort Detecting solo phrases in music using spectral and pitch-related descriptors
dc.creator.none.fl_str_mv Fuhrmann, Ferdinand
Herrera Boyer, Perfecto, 1964-
Serra, Xavier
author Fuhrmann, Ferdinand
author_facet Fuhrmann, Ferdinand
Herrera Boyer, Perfecto, 1964-
Serra, Xavier
author_role author
author2 Herrera Boyer, Perfecto, 1964-
Serra, Xavier
author2_role author
author
dc.subject.none.fl_str_mv So -- Informàtica
topic So -- Informàtica
description In this paper we present an algorithm for segmenting musical audio data. Our aim is to identify solo instrument phrases in polyphonic music. We extract relevant features from the audio to be input into our algorithm. A large corpus of audio descriptors was tested for its ability to discriminate between solo and non-solo sections, which resulted in a subset of five best features. We derived a two-stage algorithm that first creates a set of boundary candidates from local changes of these features and then classifies fixed-length segments according to the desired target classes. The output of the two stages is combined to derive the final segmentation and segment labels. Our system was trained and tested with excerpts from classical pieces and evaluated using full-length recordings, all taken from commercially available audio. We evaluated our algorithm by using precision and recall measurements for the boundary estimation and introduced new evaluation metrics from image processing for the final segmentation. Along with a resulting accuracy of 77%, we demonstrate that the selected features are discriminative for this specific task and achieve reasonable results for the segmentation problem.
publishDate 2009
dc.date.none.fl_str_mv 2009
2018
2018
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/acceptedVersion
format article
status_str acceptedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10230/34624
http://dx.doi.org/10.1080/09298210903406632
url http://hdl.handle.net/10230/34624
http://dx.doi.org/10.1080/09298210903406632
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Journal of New Music Research. 2009;38(4):343-56
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Taylor & Francis (Routledge)
publisher.none.fl_str_mv Taylor & Francis (Routledge)
dc.source.none.fl_str_mv reponame:Repositorio Digital de la UPF
instname:Universitat Pompeu Fabra
instname_str Universitat Pompeu Fabra
reponame_str Repositorio Digital de la UPF
collection Repositorio Digital de la UPF
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869417233980063744
score 15.228081

Registros relacionados