Audio Digital By Humming

INVITED PAPER

Content-Based Music Information Retrieval: Current Directions and Future Challenges
Current retrieval systems can handle tens-of-thousands of music tracks but new systems need to aim at huge online music collections that contain tens-of-millions of tracks.
By Michael A. Casey, Member IEEE , Remco Veltkamp, Masataka Goto, Marc Leman, Christophe Rhodes, and Malcolm Slaney, SeniorMember IEEE

ABSTRACT | The steep rise in music downloading over CD sales has created a major shift in the music industry away from physical media formats and towards online products and services. Music is one of the most popular types of online information and there are now hundreds of music streaming and download services operating on the World-Wide Web. Some of the music collectionsavailable are approaching the scale of ten million tracks and this has posed a major challenge for searching, retrieving, and organizing music content. Research efforts in music information retrieval have involved experts from music perception, cognition, musicology, engineering, and computer science engaged in truly interdisciplinary activity that has resulted in many proposed algorithmic andmethodological solutions to music search using content-based methods. This paper outlines the problems of content-based music information retrieval and explores the state-of-the-art methods using audio cues (e.g., query by humming, audio

fingerprinting, content-based music retrieval) and other cues (e.g., music notation and symbolic representation), and identifies some of the major challenges for thecoming years. KEYWORDS

|

Audio signal processing; content-based music

information retrieval; symbolic processing; user interfaces

I . INTRODUCTION
Music is now so readily accessible in digital form that personal collections can easily exceed the practical limits on the time we have to listen to them: ten thousand music tracks on a personal music device have a total duration ofapproximately 30 days of continuous audio. Distribution of new music recordings has become easier, prompting a huge increase in the amount of new music that is available. In 2005, there was a three-fold growth in legal music downloads and mobile phone ring tones, worth $1.1 billion worldwide, offsetting the global decline in CD sales; and in 2007, music downloads in the U.K. reached new highs [1]–[3].Traditional ways of listening to music, and methods for discovering music, such as radio broadcasts and record stores, are being replaced by personalized ways to hear and learn about music. For example, the advent of social networking Web sites, such as those reported in [4] and [5], has prompted a rapid uptake of new channels of music discovery among online communities, changing the nature of musicdissemination and forcing the major record labels to rethink their strategies.
0018-9219/$25.00 Ó 2008 IEEE

Manuscript received September 3, 2007; revised December 5, 2007. This work was supported in part by the U.K. Engineering and Physical Sciences Research Council under Grants EPSRC EP/E02274X/1 and EPSRC GR/S84750/01. M. A. Casey and C. Rhodes are with the Department of Computing, GoldsmithsCollege, University of London, SE14 6NW London, U.K. (e-mail: m.casey@gold.ac.uk; c.rhodes@gold.ac.uk). R. Veltkamp is with the Department of Information and Computing Sciences, Utrecht University, 3508TB Utrecht, The Netherlands (e-mail: Remco.Veltkamp@cs.uu.nl). M. Goto is with the National Institute of Advanced Industrial Science and Technology (AIST), Ibaraki 305-8568, Japan (e-mail:m.goto@aist.go.jp). M. Leman is with the Department of Art, Music, and Theater Sciences, Ghent University, 9000 Ghent, Belgium (e-mail: Marc.Leman@UGent.be). M. Slaney is with Yahoo! Research Inc., Santa Clara, CA 95054 USA (e-mail: malcolm@ieee.org). Digital Object Identifier: 10.1109/JPROC.2008.916370

668

Proceedings of the IEEE | Vol. 96, No. 4, April 2008

Casey et al.: Content-Based Music...