¡@

Automatic extraction of information from music data is an important and challenging issue in the field of content-based music retrieval. As part of the research effort, this study presents a technique that automatically identifies cover versions of songs specified by users. The technique enables users to search for songs with an identical tune, but performed by different singers, in different languages, genres, and so on. The proposed system takes an excerpt of the song specified by the user as input, and returns a ranked list of songs similar to the input excerpt in terms of the main melody. To handle likely discrepancies, e.g., in tempo, transposition, and accompaniment, between cover versions and the original song, methods are presented to remove the non-vocal portions of the song, extract the sung notes from the accompanied vocals, and compare the similarities between the sung note sequences. Our experiments on a database of 594 cross-lingual popular songs show the feasibility of identifying cover versions of songs for music retrieval.

Keywords: content-based music retrieval, cover version, main melody, polyphonic, accompaniments

Retrieve PDF document (200811_04.pdf)

Received February 27, 2007; revised August 17, 2007; accepted October 18, 2007.
Communicated by Jorng-Tzong Horng.
^* This paper was partially supported by the National Science Council of Taiwan, R.O.C. under grants No. NSC 93-2422-H-001-0004, NSC 94-2422-H-001-007, NSC 95-2422-H-001-008, and NSC 95-2218-E- 027-020. Part of this paper has been presented in the International Conference on Music Information Retrieval, Sept. 11-15, 2005, London, UK, Queen Mary, University of London.