One Example of How Artificial Intelligence Can Be Useful in Music FieldsSkip other details (including permanent urls, DOI, citation information)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. Please contact firstname.lastname@example.org to use this work in a way not covered by the license. :
For more information, read Michigan Publishing's access and usage policy.
Page 320 ï~~ONE EXAMPLE OF HOW ART] CAN BE USEFUL IN Francis Rousseaux LAFORIA Paris VI University, 2 place Jussieu, 75005 PARIS Act Informatique 12, rue de la Montagne Ste Genevieve, 75005 PARIS Fondation TOTAL pour la Musique 5, rue Michel Ange, 75016 PARIS ABSTRACT: Musical listening is a comple approach asks Artificial Intelligence (AI) ma musical representation, and the recognition c system may find ways of paying attention to the Particularly, this vision offers to Machine Learnt opened up by the possibility of learning fro between teachers and learners, through an educ
Page 321 ï~~the case and propositions for evolution must rely on a Thus, we need way representing musical kno Knowledge Based System; c"12, Musicologue" V evaluation grid of how the lesson is understood; sot'e evaluation provides explanations and justifications ab To realize that kind of semantic representation dynami among them: - clustering: given some knowledge field, how to c knowledge to cluster the field in simpler but mean - discrimination: given a field of concepts, how to fir - generalization: from some examples of one situati enough to describe the situation or the rule, and to Overview of our system "Le Musicologue" Regarding music and cognitive research, here are our - it provides the teacher with a new environment of different situations as he wants, and allowing the c
Page 322 ï~~Once this solution is given, a learning process will takc rule so that, when faced with problems similar to the ci to now), it will be able to propose a solution similar I Musicologue" should be able to solve problems whip capabilities (this is how to lead student progression), a is very limited: it will be developed without requiring a construction, as with a traditional Expert System. As far as student progression is concerned, some rules m; pieces, and analysis of mistakes. To produce these rules an explanation is the description of the validity domair Musicologue" will to find an explanation of the us between the concepts included in the solution. It will between the relevant and the irrelevant links in the net instantiated rules, the system must automatically vari domain of these variables by asking "clever" question (simply by turning all constants into variables) this ru sufficient conditions of the application of the instantiat of the generalized rule. An over-generalization of the considered as a set of necessary conditions for the gen
Page 323 ï~~that is, a base of educative experience which is very la a human learns. References  T.. Mitchell, S Mahadevan, L. Steinberg, "L Design", Proc. IJCAI-85, Los Angeles, pp. 573-5E  F. Rousseaux, 0. Koechlin, B. Widemann, D. I la Musique, "IA & Musique", Proc. APPLICA 88 [3) J. Chailloux, "LE LISP de 1'INRIA, Le Manue  D. Charles, "La musique et l'6criture", musiqu  S. McAdams, "Les formes du plaisir musical", I 1987.  M. Balaban, "Toward a general computer stud New York at Albany, 1984. Also, G.J. Balzano, Â~1 rriicrotonal pitch systems", Computer Music Journ "Toward a theory of formal musical languages", ra 1988.  J.E. Marie, "Sur quelques probl mes de notation  P.F. Baisnee, J.B. Barriere, 0. Koechlin, R. Ro and computer: live Derformance utilisation of the4