On 30/11/2007, Macciza Macpherson <macciza@optusnet.com.au> wrote:

From memory all the KP style ugens in the STK framework expect a
'lowest freq' at init

Makes perfect sense, yes, otherwise they might demand virtually unlimited amounts of memory. This seems to be the bit that's missing in Sitar in the ChucK version. Part of the issue might be that the STK uses constructors but ChucK doesn't support those (yet?).

Voic-form is a 4-band formant synth not actually a speech synth. good
vowel sounds though

Ok, but does that mean we couldn't get "c"/"k"/"p" etc sounds through clever usage of enveloped noise in the exciter signal? Or is there none? Some of the sounds did sound to me like a exciter/resonator system.

Check the orig STK project at http://ccrma.stanford.edu/software/stk/
index.html for more info

I've got the little book! :-) I'll read up on this one tonight when I'm back home. I'd really like ChucK to be able to say "chuck" and the STK code is quite accessible. A first step might have to be porting the code from the Ugen to native ChucK code as trying to find the right parameters for such sounds would get realy boring really fast if at every step I'd need to re-compile ChucK itself.

Yours,
Kas.