I’ve implemented a couple of new techniques in the live process. The timing of the sound triggers is now more flexible, and follows the spectral content of the source material. The individual spectral band related triggers can be adjusted separately, and the end result is livelier than before. I’ve ended utilizing a central timing mechanism based on fixed fractions rather than free oscillation. This gives the resulting music a nice fundamental structure. The base sync is usually a few beats per minute.
I’ve recorded a small demo of the system. In the demo a thunderstorm recording is transformed into music. In this example the higher frequencies produce shorter envelopes for the synthesizers in use. Enjoy a spectralization of a thunderstorm. The process happens in real time, so it never quite repeats itself, even though the storm recording would be played in loop.
The other technique of which I’ve been really fascinated is about stopping the playback of a sound to a standstill. This has been enabled by modified extreme time stretching algorithms provided by Jean-Francois Charles. Typically in a live performance I’m playing the source material between 1/100 – 1/1000 speed, and the music and other elements follow the slowly changing spectral content analyzed from the material.
I recently performed live at a library main room in Helsinki. The recording of the spectral ambient set is now available.