hi all, especially dieter ;)

remember the steinberg free filter afaik that was the first resonant
filter/graphic eq type digital processor that could 'learn' a frequency
spectrum of a sound or recording and then mimic it. if it's technically
possible then i think this idea could also be captured in a user
friendly a-100 module.

the user interface could be limited to

- a 'learn' source input and learn button
- an audio input (stereo would be great)
- an audio output
- possibly some correcting eq (hi, mid, lo)
- possibly feedback (open loop )
- an 'invert' button

actually the invert button is great in an a-100 environment. imagine you
have a complete drum sound, now you let the module learn it. next you
invert the drums-spectrum and you send some other sound like strings
and/or bass through the module and the result is that the sounds won't
be competing in the same freq. range. Could be fun i think. More room
for wild improvising without getting lost in a sonic swamp :)

To clarify: i wouldn't look for something that can clone or model
'classic hardware' or a convolution based design but something much
simpler and not too fancy like that free filter idea

greetz, Anton