This page is part of the SALIERI webpages.

[Home] · [Mail]


A grammar/case-based composition system to variate music for musicological studies in Salieri

by Wolfgang Chico Töpfer

When I first quite informally wrote that AVA was a system designed for the composition of variations on a given theme in the style of Mozart, I was only giving a very narrow idea of the concept.

The "style" of Mozart is exclusively defined by the use of a musical data base (a so-called schemes base) consisting of musical schemes ie. abstract representations of music extracted from Mozart variations such as KV 265 (well-known as "Ah,vous dirai-je maman"). This ows to my rather general focus on the complex matter of structural, human-like machine composition. Thus, in the first stage the complexity of stylistic aspects is disregarded.

The reason for the choice of Mozarts piano variations as a specimen of human composition is justified by a number of advantages that qualify them as most appropriate for the task. However, this does not mean that AVA is restricted to a schemes base of Mozart variations. The AVA /SALIERI user is free to use as many other schemes bases as RAM and disk space permit.

The main task of AVA is to assist musicologists/musical information scientists in experimenting with formal models of music in order to study the phenomenon of music composition by empirical means. The assumption is that music composition can be studied like a natural science if an experimental system like AVA is provided.

The following pages also serve as an introduction to the field of machine composition by example. This is accomplished by a description of AVA , its application to a series of downloadable experiments and the attained results .

If you still like to learn more about AVA , you may consider to download its SALIERI source code . You may also take a look at my Diploma Thesis [Chi97] in zipped Post Script (beware, it's in German!) which describes AVA in detail. An English version of my Diploma Thesis is in preparation.

Main references

[Chi97] W.Chico-Töpfer. AVA: Ein experimentelles, grammatik- und fallbasiertes Kompositionssystem in SALIERI zur automatischen Variation von Musik durch Generierung von Schemafolgen. Diplomarbeit, Oktober 1997. TUD. Fachbereich Informatik. Fachgebiet "Automatentheorie und Formale Sprachen".

[Cop91] D.Cope. Computers and musical style. The Computer Music and Digital Audio Series,Volume 6. A-R Editions. Madison, Wisconsin 1991.

[Cop96] D.Cope. Experiments in musical intelligence. The Computer Music and Digital Audio Series, Volume 12. A-R Editions. Madison, Wisconsin 1996.

[Hoo94] H.H.Hoos. Strukturorientierte Beschreibung und Erzeugung von Musik. Studienarbeit am Institut für Theoretische Informatik. TUD. Fachbereich Informatik, Juni 1994.

[Hoo97] H.H.Hoos. GUIDO: Format-Spezifikation und -Publikation. Arbeitsberichte des Fachgebietes "Automatentheorie und Formale Sprachen". Fachbereich Informatik, September 1997.

[Jac83] F.Lerdahl, R.Jackendoff. A Generative Theory of Tonal Music. The MIT Press series on cognitive theory and mental representation. MIT Press, 1983.

[Jac93] F.Lerdahl, R.Jackendoff. An Overview of Hierarchical Structure in Music. First publication in: Music Perception, 1983. Current source: Machine Models of Music. MIT Press.

[Kol93] J.Kolodner. Case-based reasoning. Morgan Kaufmann 1993.

[Nar90] E.Narmour. The Analysis and cognition of basic melodic structures: the implication-realization model. University of Chicago Press 1990.

[Home] · [Mail]
© sic!systems, designed by hh; last update 1997/12/31