Please use this identifier to cite or link to this item: http://hdl.handle.net/10316/15404
DC FieldValueLanguage
dc.contributor.authorGonçalo Oliveira, Hugo R.-
dc.contributor.authorCardoso, F. Amílcar-
dc.contributor.authorPereira, Francisco C.-
dc.date.accessioned2011-06-27T11:41:43Z-
dc.date.available2011-06-27T11:41:43Z-
dc.date.issued2007-06-
dc.identifier.urihttp://hdl.handle.net/10316/15404-
dc.description.abstractThis paper is about an ongoing project, Tra-la-Lyrics which aims to create a computer program capable of gen- erating lyrics for given melodies. As a preliminary phase, the correlations between words and rhythm in a song’s lyrics was studied and some heuristics were achieved. Al- gorithms for syllabic division and syllabic stress identifi- cation on a word were implemented. A method for cal- culating the strength of each beat of a melody was also used. The actual system’s architecture is described. To get the contents of the lyrics we’ve used a relational database where we could find not only the words, but also their grammatical category and some morphological related at- tributes. Some ongoing work is also referred and some examples of the currently possible outputs of the system are presented. Conclusions and possible further work are finally discussed.por
dc.description.sponsorshipCISUCpor
dc.language.isoporpor
dc.publisherA. Cardoso and G. Wigginspor
dc.rightsopenAccesspor
dc.titleTra-la-lyrics: an approach to generate text based on rhythmpor
dc.typepreprintpor
degois.publication.locationLondon, UKpor
degois.publication.titleProceedings of the 4th. International Joint Workshop on Computational Creativitypor
dc.peerreviewedYespor
item.fulltextCom Texto completo-
item.grantfulltextopen-
item.languageiso639-1pt-
Appears in Collections:FCTUC Eng.Informática - Artigos em Livros de Actas
Files in This Item:
File Description SizeFormat
GoncaloOliveiraetalIJWCC2007-reformatted.pdf260.55 kBAdobe PDFView/Open
Show simple item record

Page view(s) 50

330
checked on Nov 6, 2019

Download(s)

62
checked on Nov 6, 2019

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.