Extending Csound Using Scripting Languages

Csound orchestras and scores are in the form of pure ASCII text files. Thus it’s relatively easy to generate either separate .orc and .sco files or, more likely, unified .csd files, using another programming language. The other language may also be able to invoke Csound directly, causing the generated file to play. Using Csound in this way is, of necessity, more complex than using Csound by itself, as you’ll have to master two separate programming languages. The main reason to use this approach is because you can generate massive scores that would be impractical to produce by entering one note at a time in a text editor.

A granular synthesis texture, for instance, could be produced in which each sound grain was a separate event in the score, each event making use of a dozen or more p-fields. Thousands of score events might be needed for each minute of sound output. Music requiring complex, precisely calculated polyrhythms would also be a good candidate for this approach.

Various scripting languages can be used for this purpose. In Issue 5 of the Csound Journal, in an article on using Perl with Csound, Jacob Joaquin mentions, in passing, that JavaScript, Lisp, PHP, Python, and Ruby are all options. “I recommend avoiding using languages such as C and Java,” Joaquin adds, “as they are particularly unwieldy in processing and managing text compared to the languages I’ve just mentioned.”