posted 12 March 2006 06:09            

WavesetAnagrams.kym

inspired by a poet who used the anagram*-technique to create variation within his poetry, I'd like to transfer this idea to audio-domain, utilizing again Wisharts waveset-technique - it makes sense here as a flexible criterium for subdividing a signal.
You'll find attached all the sub-patches and their combination in one big 'ideal' patch**, doesn't work completely yet...

Problems so far (which I couldn't solve on my own):
1. The calculated waveset-timing/durations (sub-patch 'ws triggers and duration'): they have to be 'distributed' to the individual wavesets (sample modules), to their start- and stop times and to their pseudogates (FGs). But how? How just supply one value per waveset (sample) from the Triggered S&H sound?
2. Random order sequences of the wavesets (sub-patch 'random sequences'): the simple example works** because I worked out a table of possible combinations, which I then implemented in form of arrays into the delay scale fields.
But firstly, this is not practical for longer sequences with lots of wavesets - the tables would get too enormeous to be implemented in advance.
Secondly, this is not a live-solution, since you can't anticipate the wavesets and their durations.
Thirdly, and that brings me to the next issue, thus
3. :-) This should be a realtime tool for live interaction, with a flexible (adapting to the music) rec-time and waveset-size (both being controllable). The number of wavesets can vary from 2 to for ex. 20, from recording to recording. And that means that the number of sample-modules in the patch should be variable, depending on the number of wavesets you created; and that depends again on the size of the recording and the size of the waveset.
How can you make this variable within this patch? With a script? Can somebody help me out with this?
Thanks in advance.
Best, Florian