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


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

This page now includes a link (through the icon) to the full text of the final version of the conference papers. These texts are © IRCAM – Centre Pompidou in this form. Authors have retained the rights to their original texts.

The printed version of the proceedings, which also contains introductions, the abstracts, a table of contents and the author index, can be ordered here.


This special session, consisting of invited distinguished speakers, will discuss the importance and variety of metadata for music information retrieval. Whether derived automatically from contents (audio, score...) by extracting a variety of features (samples, notes, melody, form, lyrics...), or produced manually by specialists (e.g., as in libraries), supplemented with such mechanisms as authority control, and put in relation with musical documents and documentation, very powerful music searching tools may be developed. This session will also address the issues of creating, organizing, managing and using potentially very large collections of metadata, including the technical and economical/marketing aspects.

This session will be held on the afternoon of Wednesday, October 16, 2002, and will be open to Résonances visitors. The participants and the topics of their talks are:

Chris Barlas

“Beating Babel - Identification, Metadata and Rights”

The development of rights trading for music over networks depends, crucially, on persistent identification and the availability of machine readable descriptive and rights metadata that is interoperable. However, with the proliferation of metadata schemas and identification systems, the future is potentially very confusing.

The presentation describes an approach to this subject, beginning with the <indecs> analysis, developed as part of the Info2000 programme. Now widely accepted in the content industries, the <indecs> analysis has informed a number of initiatives, including the MPEG-21 Rights Data Dictionary and the GRID project (the Global Release Identifier) for the release of music on-line.

The presentation also describes the work being carried out by the Contecs:DD Consortium, based on the original <indecs> work, which has been adopted as the baseline technology for the MPEG-21 Rights Data Dictionary, currently at Committee Draft stage.

Leonardo Chiariglione

“Technology and art – putting things in context”

A set of standard technologies already developed or under development by the Moving Picture Experts Group (MPEG) promises to bring back to authors the control of their works. The technologies are those of Content Representation, Digital Item Declaration, Interoperable Intellectual Property Management and Protection, and Metadata.

Dave Datta

“Managing Metadata”

The All Media Guide (AMG) is a technology company that maintains the world’s largest database of metadata relating to the entertainment industries. This document describes some of the goals of AMG, the issues uncovered during the evolution of our databases, and discusses some of the implementations we have chosen.

Harriette Hemmasi

“Why Not MARC?”

Traditional library cataloging records in the United States, based on AACR2R cataloging rules and MARC standards, constitute a solid foundation for many of the descriptive metadata elements needed for searching and retrieving works of music.  However, there are significant weaknesses associated with these records and the online environment in which they live as users seek access to digitized representations of music.  While music metadata in the library catalog records offer less than a perfect solution, they can and should have an important role in the total solution.  Variations2, the Indiana University Digital Music Library, builds on the advantages of AACR2R and MARC and offers a domain-specific data model and search environment that address many of the  identified problems. 

Eric Scheirer

“About this Business of Metadata”

A brief discussion presents some of the opportunities and challenges involved with creating metadata-centric businesses that bring Music Information Retrieval technologies to the marketplace.  In particular, two related difficulties -- that of the difficulty of proving incremental value for new metadata systems, and that of the relative influidity of the marketplace for MIR -- are highlighted.  Potential directions for resolving these issues are also discussed.


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