Phasing click track
Steady and phasing click track
While preparing a real violin version of Violin Phase, I was able to experiment with different microphone setups, all of which have an audible effect on the recording quality. The following were recorded on the CCRMA stage.
Cartiods with some overlap. Violin ~2 ft away.
Cartiods with some overlap. Violin ~1 ft away.
Cartiods with significant overlap. Violin ~2 ft away.
Cartiods with significant overlap. Violin ~1 ft away.
Subcartiods with significant overlap. Violin ~1 ft away.
Sony recorder. (Recorded in Studio D)
Here is a link to Esteban's research relating to violin physical models.
The program can now take any simple input pitches and rhythm and accent the important beats by increasing the volume slightly.
I've been laying out my program so that it is simple to use. Public class functions will allow the user to specify some details about the input music, but the output will automatically be musical.
Ideally, I could use this class to make the tracks to Violin Phase.
Can I specify enough parameters to capture salient violin techniques, i.e. those that would make a phrase sound stringy? The bare minimum is a sawtooth with a 20 ms envelope.
For simplification, I will feed the program only single phrases. See code for the parameters that have been considered, but not yet implemented.
I have rethought my project. The previous project required learning and adding functionality to a new tool. I am not so interested in making the tool, but in getting the results. Still involving analysis, I want to instead make computer generated sounds sound more musical. The idea is for the computer to "understand" the music such that it can intelligently enhance what it generates.
I have been reading Lewin's article on phemonenology (Lewin, David. Music Theory, Phenomenology, and Modes of Perception. Music Perception: An Interdisciplinary Journal, Vol. 3 No. 4, University of California Press: 1986, pp. 327-392). This has made me reconsider my previous understanding of his mathematical theory. In GMIT, I assumed his pointing out of intersting mathematical operations was not driven by any goal other than to "decompose" the music. After reading part of this paper, which predates the book, I have realized that his motivation was perceptual in nature.
It would be interesting to analyze a work à la Lewin knowing that it is meant to fit into a broader context. It was suggested that the process of finding relevant theoretical patterns in music could be automated. Will look into PWJL and CCARH.
Here's a related link on Neo-Riemannian theory.