Re: An idea for an entirely new effect -- warning, long post

Federico wrote:

 > It models a signal by detecting partials and representing their time 
evolution by sine waves

Thanks for the information, but here this is not necessary. The 
frequency-shifting part can be done entirely in the time domain. What is 
necessary, however, is to have a sufficiently long buffer for the filter 
that does the Hilbert transform.
It's a bit like reverberating a signal that has already been 
reverberated. Imagine that you would run your signal through a reverb 
effect and that you would feed your result back into the input, delayed 
by, let's say, a few seconds. So rather than having something like a 
repeating echo where every repetion sounds softer than the previous one, 
you have repeating reverberation where each repetition stresses the 
reverb characteristics more than the previous one.
This frequency-shifting stuff is similar, except that the delays can be 
as small as a fraction of a millisecond.
Once we can answer the question how we could achieve the repeating 
reverb that I've just described, we should be able to answer the 
question how we can do the repeated frequency shifter.

Petr


-- 
Tento e-mail byl zkontrolován na viry programem AVG.
http://www.avg.cz



_______________________________________________
audacity-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/audacity-devel