Change search
ReferencesLink to record
Permanent link

Direct link
Singer Identification in Rembetiko Music
Institute of Computer Science, FORTH, Greece.ORCID iD: 0000-0003-1679-6018
2007 (English)In: Proceedings of the Conference on Sound and Music Computing (SMC), Sound and music Computing network , 2007, 326-329 p.Conference paper (Refereed)
Abstract [en]

In this paper, the problem of the automatic identification of a singer is investigated using methods known from speaker identification. Ways for using world models are presented and the usage of Cepstral Mean Subtraction (CMS) is evaluated. In order to minimize the difference due to musical style we use a novel data set, consisting of samples from greekRembetiko music, being very similar in style. The data set also explores for the first time the influence of the recording quality, by including many historical gramophone recordings. Experimental evaluations show the benefits of world models for frame selection and CMS, resulting in an average classification accuracy of about 81% among 21 different singers.

Place, publisher, year, edition, pages
Sound and music Computing network , 2007. 326-329 p.
Keyword [en]
Artist Identification; Gaussian Mixture Model; Music Information Retrieval
National Category
Media Engineering
Research subject
Computer Science; Media Technology; Speech and Music Communication
Identifiers
URN: urn:nbn:se:kth:diva-193766ScopusID: 2-s2.0-84905189665OAI: oai:DiVA.org:kth-193766DiVA: diva2:1040356
Conference
Conference on Sound and Music Computing (SMC)
Note

QC 20161031

Available from: 2016-10-27 Created: 2016-10-10 Last updated: 2016-11-11Bibliographically approved

Open Access in DiVA

fulltext(200 kB)6 downloads
File information
File name FULLTEXT01.pdfFile size 200 kBChecksum SHA-512
e51ba7604f2e93c6e6d4e7c99ee3f8bb99dc8e26742b5a078ef6c372dba0852b926c3c4c172e29a41e370a0562cadeddfa5b6e6a277882f081f076791a3f8e20
Type fulltextMimetype application/pdf

Scopus

Search in DiVA

By author/editor
Holzapfel, André
Media Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 6 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 1 hits
ReferencesLink to record
Permanent link

Direct link