what would you be your preferred to turn a dynamic audio signal, preferably from a [groove] object into a steady-level signal,
without clipping or distortion,
eventually running with an external & repetetive adsr applied?
i tried to compress up to the amount of brickwall limiting, but that doesn’t give me the desired effect of turning a dynamic signal into a highly congruent ambience, without the use of reverb or delay.
thank you to the one inclined, and for any hint pointing me somewhere!
> i tried to compress up to the amount of brickwall limiting, but that
> doesn’t give me teh deireed effect of turning a dynamic signal into a
> highly congruent ambience, without the use of reverb or delay.
You could try to do some spectral smearing over several fft frames…
There was a thread about such a technique some time ago. As you want to
get rid of attacks, just destroy them…. ;-)
(Ibook crashed yesterday – momentary stop in further developemnt, nevertheless:)
thanx very much for the suggestion. I had looked into an FFT solution, but generally found the resulting sound quality not convincing, which seems to be a well-known drawback of FFT in general.
So instead i am trying to apply rather typical sound engineering methods to the task of "ambientification".
the [lbyl] object from the cnmat library turned out to be quite useful, as it kills sudden spikes easily.
[average~] then does some good things as well, although distortion does not permit the use of the absolute and rms mode in this matter. i am hoping to supply some half-working result by doing a frequency band split after adc and treat several different bands with different methods, not included delays or reverbs, as i want to keep the resulting ambience in it’s own room setup as far as possible.
unortunately, as i pointed out in brackets above, i am hindered by the breakdown of my osx-running computer momentarily, will nevertheless post positive achievements anytime i get back to having a proper work platform (soon!)