Vol. 26 Issue 2 Reviews
Wedelmusic: First International Conference on Web Delivering of Music
Florence, Italy, 23-24 November 2001

Reviewed by Denis Baggi
Manno, Switzerland

The Wedelmusic conference, which received its name from the WeDelMusic (Web DELivery of MUSIC) project, is the first of a series which is intended to deal with the distribution of music, including protection and fruition, transaction models, modeling, streaming, new media, conversion, and cultural heritage. The conference took place in Florence on the 23rd and 24th of November 2001, with about 60 participants from all over the world, including Australia, USA, Canada, Singapore, Germany, France, Austria, Denmark, England, Switzerland, and of course, Italy.

The level of the contributions was generally very high, always at the scientific and technological edge of current research, and the organization was perfect, with well-working audio and video devices. The interaction among participants was likewise very effective and provided many opportunities for exchanges across the several areas represented.

Here follows a brief description of each contribution, with reference to the page numbers in the Proceedings (available from IEEE Computer Society Press, P. O. Box 3014, Los Alamitos, California 90720-1314, USA; electronic mail cs.books@computer.org).

23 November 2001
After the Greeting by Conference Chair Paolo Nesi of the University of Florence, the first lecture was by Keynote Speaker Denis Baggi of the Institute for Applied Computer Science and Industrial Technology (CIMSI) of the Professional University of Southern Switzerland. This talk, “Understanding Jazz: The Structures of Swing” (p. 2), was not an academic lecture but a description, with examples, of the speaker’s multimedia system (book + CD-ROM) to explain the inner working of jazz improvisation. The originality of the approach lies in the fact that every musical example can be heard immediately, and that jazz history is treated in function of improvisational structures that traverse all periods. A few dozens examples were presented, with a demonstration by the speaker on his soprano saxophone.

Mr. Baggi also announced the existence of a new Standards project (IEEE Project Authorization Request 1599, dated September 28, 2001), concerning the “Definition of a Commonly Acceptable Musical Application Using the XML Language.” The interest of Intelligent Manufacturing Systems (www.ims.org) in financing this effort was also noted.

The first session was on Music Protection. Michael Arnold, of the Fraunhofer Institute in Darmstadt, presented “Blind Detection of Multiple Audio Watermarks” (p. 4), a system of algorithms to enable blind detection of watermarks, including multiple ones. This was followed by Martin Schmucker, of the same institute, presenting “High Capacity Information Hiding in Music Scores” (p. 12), techniques to add watermarks to a score. He also presented “Using Musical Features for Watermarking Music Scores” (p. 20), on how to exploit musical features such as beams, slurs, and ties to embed watermarks. “Watermarking Music Scores While Printing” (p. 28) was introduced by Marius Bogdan Spinu of the University of Florence, a new technique for watermarking music while printing music sheets. Franco Bartolini, of the same university, presented “Watermarking-Based Copyright Protection of Internet-Delivered Multimedia” (p. 36), a proposal for an Electronic Copyright Management System to enable protection in open networks. Finally, “Content Protection and Usage Control For Digital Music” (p. 46), was delivered by Yongwei Zhu of the Kent Ridge Digital Lab in Singapore. This work concerns a watermarking scheme for both compressed music and MIDI format.

The next session, Music Editing And Recognition, started with “Automatic Formatting of Music Sheets” (p. 170), presented by Riccardo della Santa of the University of Florence, a set of conditions and rules for music formatting formalized for execution by the Music Intelligence Formatting Language system. George Giannopoulos, of the University of Athens, discussed “Music Editors for Visually Impaired People: User Interface Specifications and System Design” (p. 178), user-interface specifications and software systems for the electronic editing of music by blind people. Lastly, Ivan Bruno, again of the University of Florence, presented “Optical Music Sheet Segmentation” (p. 183), a system that recognizes and extracts basic symbols from a musical score.

The last session of the day was on Applications. “The European Music Navigator: To Build Bridges Between Local, National and Global Cultures, Using the Ontology-Based Search Technology MelvilTM” (p. 192) was presented by Peter Rantasa of the International Association of Music Information Centers in Austria. Kai Renz, of the University of Darmstadt, presented “WEB Delivery of Music Using the GUIDO NoteServer” (p. 193), an online service to deliver graphical images of a music score. The session ended with an exhaustive demonstration of the WeDelMusic music editor by Mr. Nesi, a system which deals with all sorts of music encodings: score, images, texts, and the like (see also the article on p. 79 described below).

November 24, 2001
The next day started with the session Music Modeling and Distribution. Andreas Kornstädt, of the Center for Computer Assisted Research in the Humanities, Stanford, discussed “Data Models for Virtual Distribution of Musical Scores” (p. 62), the MuseData model for score archiving and distribution. “Music Information Description by Mark-Up Languages within DB-Web Applications” (p. 71), presented by Maurizio Longari of the Laboratory for Musical Informatics of the University of Milan, was a review of past standards, and focused on the use of XML.

“ WeDelMUSIC Format: an XML Music Notation Format for Emerging Applications” (p. 79) by Pierfrancesco Bellini of the University of Florence, described the WeDelMusic project with examples of XML encoding.

The session on Audio Fruition and Manipulation began with “The Neural Network Model of Music Cognition ARTIST and Applications for the WWW” (p. 155), by Frédéric Piat from Paris. It described a Neural Network capable of simulating high-level perceptual and human cognitive abilities with regard to human musical preferences. In his paper, “Expressive Morphing for Interactive Performance of Musical Scores” (p. 116), Antonio Rodà, of the University of Padova, described a system to process in real time the expressive character of the rendition of a score. Max Mühlhäuser, of the University of Linz, Austria, presented “GlobeMusic: The Internet Scale of eMusic Making” (p. 131), an Internet system for computer-human interaction, individual eMusic-making, cooperative eMusic-making, and the like. Florian Pestoni, of the IBM Almaden Research Center near San Jose, described “KARC: Radio Research” (p. 139), a system to select a radio station in function of a user’s musical tastes. “Real Time Musical Events Streaming over Internet” (p. 147), introduced by Dominique Fober, of Grame Computer Music Research Laboratory, Lyon, is a protocol for streaming real-time events over the Internet. Finally, “The Recording Studio that Spanned a Continent” (p. 161) was the contribution of Jeremy Cooperstock, of McGill University, Montreal, describing a real-time concert played in Montreal and heard in Los Angeles via broad-band Internet. His presentation also touched on related issues such as real-time “jamming” over the World Wide Web.

The next session, Cultural Heritage, consisted of one contribution, “Saving the Multimedia Musical Heritage of Teatro Alla Scala for Querying in a Web-Oriented Environment” (p. 52), by Goffredo Haus of the Laboratory for Musical Informatics at the University of Milan. He described the rescuing, processing, archiving, and structuring of the documents—audio, video, and other—of La Scala, accessible through a multimedia database.

Music Analysis and Classification was the final session of the conference. “Classification of Melodies by Composer with Hidden Markov Models” (p. 88), by Emanuele Pollastri of the Laboratory for Musical Informatics at the University of Milan, focuses on a Hidden Markov model to abstract the style of a composer. “Content-based Identification of Audio Titles on the Internet” (p. 96) was presented by Helmut Neuschmied of the Institute of Information Systems Joanneum Research in Graz, Austria, and concerns a system based on Audio DNA, from which the title of a piece can be extracted. François Pachet, from Sony, Paris, discussed his “Musical Data Mining for Electronic Music Distribution” (p. 101), a method of classification for music. And, Jérome Barthélemy, of the Institut de Recherche et Coordination Acoustique/Musique, Paris, presented “Similarity on Computational Music: a Musicologist’s Approach” (p. 107), regarding methods for musical analysis.

The WeDelMusic Conference concluded on Saturday, 24 November. Other related activities, such as a session on Electronic Music Publishing, continued at the Villa Strozzi in Florence, home to Tempo Reale, active in computer music.