I'm back (sorry, it takes me a long time to finish things, too many hobbies and not enough "free" time).
I implemented the freeverb schroeder AP section and fed it with a short sample then cross-faded the input to be 100% feedback from the input. The feedback was delayed the length of the sample. You can see/download the code here: https://github.com/ssj71/infamousPlugins/blob/33e71229025920a1a1a2521a3dcbae0869034292/src/stuck/stuck.c I played quite a bit with the fade in of the sample, the sample length, and the crossfade time between the input sample and the feedback but I was not able to get satisfactory results. With long samples (>100ms) there was potential that with a little compression/envelope removal it could have worked but that requires a long stable note/chord which isn't really going to work for my application. If anyone has the time to review the code and suggest improvements that would be awesome, but already you've all been really helpful. Thanks for that. I played with the freeverb plugin and was impressed with the quality of the freeze function, but it seems that the comb filters are necessary for it to work well. So I'm trying to decide if my next step is to add the comb filters and deal with the coloration (with 8 of them in parallel it actually doesn't seem to have too deep of nulls) or to move on to the phase vocoder method. I suppose the logical next step is first the combs, but I thought I'd check in first just in case any are interested. Thanks again, _spencer On Sun, Oct 2, 2016 at 7:20 AM, Emanuel Landeholm <emanuel.landeh...@gmail.com> wrote: > Fabian (and list), > > This looks reality interesting. Are you estimating PSD FIR-coeffs using that > Burg algorithm? I have seen something similar that produces second order > sections for IIR. I believe it's called CELP. > > /Emanuel > > > _______________________________________________ > dupswapdrop: music-dsp mailing list > music-dsp@music.columbia.edu > https://lists.columbia.edu/mailman/listinfo/music-dsp _______________________________________________ dupswapdrop: music-dsp mailing list music-dsp@music.columbia.edu https://lists.columbia.edu/mailman/listinfo/music-dsp
