The music machine: Selected readings from “computer music journal”

Computer applications to music cover a very large field. This book is a collection of articles, previously published in Computer Music Journal, one of the most respected journals in the field, which deal with some of the possible applications of automated tools in music. The book, edited by the one-time editor of the journal, consists of 54 contributions divided into eight chapters: “Interviews,” “Composition,” “The Midi Interface,” “Music Software,” “Synthesis and Signal Processing,” “ Signal Processing Hardware” and “Music and Artificial Intelligence.” In the first chapter, computer musicians are interviewed on their experience of using computers to produce sound or musical compositions and in the second, several composers describe their compositional experiments. The next four chapters focus on both hardware and software developments, mainly to process signals or to manipulate musical data. A few treat music printing or applications of stochastic processes to music or describe commercial packages. The last chapter, on AI and music, is surely the most interesting for people working in computational linguistics because it concerns problems, like formalizing knowledge, theory testing and so on, which linguistics and music share. The short but very interesting article by John Rothgeb, “Simulating Musical Skills by Digital Computer,” focuses on the use of the computer to analyze what the theories of unfigured bass lack. He implemented rules extracted from theoretical attempts to harmonize the bass part, but the resulting failures demonstrated the lack of explanatory power of the theory. Therefore, the computer can be used to investigate musical theories by implementing them and then redefining or improving them by studying the results obtained from observation of how the implementations work out. Another essay of interest in this chapter is John Rahn’s “On Some Computational Models of Music Theory” in which some theoretical and practical issues raised by other authors are discussed and criticized. Of interest from the computational linguist’s point of view would be the discussion and criticism of the possible relationship, proposed by Meehan (1980), between Schank’s Conceptual Dependency Theory and the theory of implication/realization worked