It has been quite some time since I had new music to post, but not for lack of working on it. Instead, I have had several projects going on at once and only just finished these two. There are several other pieces in the pipeline, some like these, some not. Other priorities dictate my availability to work on music, but I hope the next post will not be as long delayed as this.
One of the things I have been especially interested in of late is expression, how it inheres in synthetic music differently from live performance, and how to bridge the gap between being a performer of music and a constructor, as it were, while still keeping the breath in the music. As an exercise, I began toying with simple pieces, the Bach Two-Part Inventions, for instance, and exploring how I might get the music to feel as close to live as possible. Here is the result of my first study:
Dynamic is my primary mode of expression here, with no change in tempo until the end and very little change in length of note (legato, marcato, etc.). This was intentional; it allowed me to focus on one aspect of synthetic expression (velocity) so as to explore its effects as thoroughly as I could. Still, I am pleased with the result, not least because of the quality of the sample. In this simple, two-part work, it is difficult to tell from a real piano.
The second piece is the same Invention, but now voiced synthetically. Building on the foundation of expression I developed in the first piece, I chose voice development as the next task. Of course, this skill is a lifetime of study, but I have done very little development from scratch and felt that this minimally-structured piece would be a good venue for some real timbral exploration. Thus:
Crystal is a surprisingly sophisticated free softsynth (also available for iPad and iPhone), allowing each voice to be comprised of up to three separate voices or channels, all of which can be filtered and modified separately. Initially, I built a voice comprised of a noisy, quasi-vocal sample, a relatively clean saw square, and a noise channel; I liked the edgy, forceful, almost twisted sound that resulted. However, it bludgeoned the listener a bit to play the Invention beginning to end using only that voice, so I did two things: first I played with the equalization of the right- and left-hand parts, emphasizing the high and low ends of their spectra respectively (also a new tool for me). Second, I parsed out the voice channel from the saw square and noise channels. This gave me four related but distinct voices to play with and allowed me to build the voices as the piece progressed. The result is what you hear above.
At this stage, I feel like I am just beginning to tap into the potential of this medium. I can't describe how delightful it is to be exploring this immense realm, something I've dreamt about since I first heard Wendy Carlos. I have much yet to learn, but the task is a pleasurable one. As always, I hope you enjoy the fruits of my expedition as much as I do.