posted 01 September 2004 06:18            

I am thinking about a new live performance piece, and not sure how to approach it, so thought I would aks for some advice.

I would like to write a new piece for flute and Kyma that takes the live input and slowly expands or contracts the spectral content of the input sound so as to make Xenakis like glissandi - both thick and thin layers of shimmering sound from simple pitch input from the flute - the piece would I hope be a meditaive, and engaging multichannel environment.

So I am wondering where to start with the compressor/expandor ideas that work in the spectral domain and that can be scripted so perhaps the various sepctra open and close at different rates, and so that a number of these processors could be active at one - one per input pitch, up to a reasonable limit - so I guess there would need to be a way in which the process unit (sound) would be available to take input and unavailable for new input while processing. Idea anyone??

posted 02 September 2004 11:17            

live input--> LiveSpectralAnalysis-->SpectrumModifier-->OscillatorBank