The availability of large music repositories poses challenging research problems. Among all, content-based identification is gaining increasing interest because it can provide new tools for easy music access and retrieval. In this paper, we propose a methodology for cover identification, in particular focusing on pop and rock genres, which is also motivated by the large amount of user-generated music content that is available online. Identification is based on the use of chroma features, which are indexed to achieve high scalability on the size of the collection. To this end, the approach does not exploit techniques that have a complexity linear with the size of the collection, such as alignment between the query and the documents or other forms of direct comparison. Evaluation results are provided to show the validity of the approach in two different scenarios, either as an independent cover identification system or as a clustering technique for a query-based partitioning of the music collection.

Content-based Cover Song Identification in Music Digital Libraries

MIOTTO, RICCARDO;MONTECCHIO, NICOLA;ORIO, NICOLA
2010

Abstract

The availability of large music repositories poses challenging research problems. Among all, content-based identification is gaining increasing interest because it can provide new tools for easy music access and retrieval. In this paper, we propose a methodology for cover identification, in particular focusing on pop and rock genres, which is also motivated by the large amount of user-generated music content that is available online. Identification is based on the use of chroma features, which are indexed to achieve high scalability on the size of the collection. To this end, the approach does not exploit techniques that have a complexity linear with the size of the collection, such as alignment between the query and the documents or other forms of direct comparison. Evaluation results are provided to show the validity of the approach in two different scenarios, either as an independent cover identification system or as a clustering technique for a query-based partitioning of the music collection.
2010
Digital Libraries
9783642158490
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11577/2473919
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact