rasmus ekman wrote:

> Erik Spjut wrote:
> >
> > lpreson and lpfreson interpolate between frames but unless kr=sr you'll get
> > a buzz or noise at kr. I suspect that's your burble.
>
> Huh? Isn't the burble a rather common side-effect of FFT processing:
> It comes from cross-fading the analysis/resynthesis windows, when
> the sine-waves in each frame (window) are faded up/out.
>
> 
> Standard FFT is old, primitive and grotty and should be replaced by
> multi-band sine-wave analysis (not the wavelet sine packets used for
> data reduction) right now. I remain amazed that maths-proficient
> computer musicians refuse to shower themselves in glory by developing
> some such highly useful tools!
> (spectrum is a start though)
> 
>
>         re

First:
    LPC (lpreson/lpfreson) != FFT. It is a filter matching technique.
Next:
>multi-band sine-wave analysis
    = a good general description of FTT!?
Perhaps you are refering to the family of  "multi-rate analysis", of which
"wavlet" analysis is a member.
And they are *all* prone to windowing artifacts *except* "pitch-synchronous"
'wavelet' analysis, which is great, but requires several passes, the first
determining the
raw pitch of the analysed file. Of course this doesn't work when you have a
polyphonic
sample to analyse....generally, you just can't get away from windowing
artifacts.
Lastly:
   Erik is maybe correct, AFAIK, with the LPC "burble" effect. Using the
analysis type
that keeps analysis data in polar form, and then converts to actual coefficients
should help.
If all interpolation between filter frames is done in polar (descriptive) terms,
then converted,
rather than interpolating actual filter cooeffs., then the results are better.
Perhaps, tho, the burble is more to o with an unstable pitch track result? With
Paul Lansky's
LPC code, I *always* had to hand edit the pitchtrack values to end up with a
listenable result.
I had a UNIX cmdline utility to do this once...sorry, no Weendoze, but
*this* would be a great boon to Csound: a general purpose LPC analysis data
editor.
Heck, how about a gen. purpose *all* analysis data editor?
Gee, if I only had one of those tenure track job thingies, rather than being
thrown out into
the cold cruel world to make a living I think I could do that....
well, after I retire in ten years, maybe.

CharlieB