Analog Filters -> Digital

[Brian Towles]

Sean Costello wrote:

... snip snip ...

> One other thing:  I think that computers are getting to the point where
> you CAN implement an FFT-based vocoder in real time.  I think that the
> basic idea is to take an FFT of the two signals, and transfer the
> modulus (amplitude) of each frequency bin in the signal that is the
> modulating signal to the corresponding frequency bin of the "source"
> signal.  It would be very cool, no?
>
> Later,
>
> Sean Costello

thanks for the info .... yeah, jumping into the csound source doesn't sound too
easy, but it might be worth it.

as for ffts: i think you could run into problems related to the delay
assocaited with gathering the data points for an FFT, not actually doing the
computation.  for example, it might take 5ms of samples to detect the low-end
of frequencies you would be interested in for vocoding ( in the 100Hz range ) -
this limits the speed of update for all the filters. one solution to this
problem is to use overlapped fft windows, but then computation speed starts
becoming an issue.  these problems occur in the modulation analysis.  the
delays caused by the fft-ifft pair for the output would be even worse.

i believe that discrete filters will give much better time response to
modulation changes and allow for a faster output path, without the problems
associated with the ifft, but i guess i'll see :)

brian