General information

Important dates

Submission information

Conference program

     Detailed program

     Keynote speakers

     Special Metadata Session

Accepted papers

Accepted posters/short papers



     Professional visit

     Social events

     Closing remarks


Conference venue

Conference committee

Photos from the conference

Conference proceedings


Hosted by


Sponsors and partners

Mairie de Paris

IU seal, red on white, small


ISMIR 2002
3rd International Conference on
Music Information Retrieval

IRCAM – Centre Pompidou
Paris, France
October 13-17, 2002


As ISMIR is drawing to its end, let me first remind you that there are three exciting panel sessions tomorrow. The multiplicity of interesting topics and speakers has been the reason for parallelizing some of the ISMIR program, to the chagrin of some who would have liked to be everywhere (and also visit Paris), and this would really be the only morally justified argument for research in human cloning.

Speaking of research – which, in French, is the same word than the one used for “search” – let me share very briefly a few thoughts about ISMIR. Speaking at the “meta” level, these are, in my opinion, some of the challenges and stakes that we will have to face:

·         Scientific ones: these issues are indeed well addressed, in many ways, here. We will have to take particular care in reflecting and holding the right balance between cutting edge research and innovation, on the one hand, and continuing work on existing paradigms and improving current technology on the other hand. Innovation, the hardest thing to decree, might have better chances to occur if we do indeed strive on multidisciplinarity without losing our identity, encourage it, and improve our communication skills not within our own specialty, but between specialists of different branches, from computer science and digital signal processing to information and library sciences, perception, musicology, law, business and many others.

·         Economical: there is a tension between the (shorter term) interests of “industry” and “the [entertainment] market” on the one hand and the intellectual/longer term goals of “academic research”. Who would have imagined the “real life” applications of Fermat’s Theorem or Gallois’ Field Theory (after all my first academic training was in number theory)? So one should definitely not dismiss what appears at first to be an idle pursuit. Rather than ignore this tension between industry and academia, both “camps” should turn it into something positive (in the same way we saw earlier today the ways in which “web metadata” started influencing “library metadata” and conversely).

·         The legal aspect was quite absent from ISMIR this year, but as you may be aware, these are very complex issues when “merely” considering music on the Internet, say (I don’t believe that it will ever become a “free” fluid – is there a free fluid? Fluids are becoming more and more expensive. But as predictions go…). But when considering derived elements (music excerpts, incipits, summaries and so on) they may also be subject to complicated legal issues.

·         Finally, one should not overlook the multicultural and social aspects of our domain. Music is not just 17th-19th century Western European music, but is many other things, as those of you who came to yesterday’s concert may have heard, and as the ethnomusicologist Simha Arom will most probably broach upon tomorrow at the Similarities panel.

Having thanked many people and organizations who helped in the making of ISMIR 2002, I’d like to thank, this time, in particular, all of you: the authors and the public, and hope you continue developing this field we all like.

The ISMIR 2002 Web pages will be regularly updated
to include program content and schedule