Hi, guys!
I'm trying to work on some binaural spacialization. And I'm aware there's already some externals like vbap~ or space~, but they don't quite work for me. However I came across the CW_binaural~ which completely kicks ass. But I'm trying to work this on myself without any use of externals.
So, theory dictates that convolution of any sound with an HRIR converts the sound as if it was being heard from the HRIR source location. But I'm not getting it completely right. btw, im using the HRIR's from the listen database (http://recherche.ircam.fr/equipes/salles/listen/download.html)
let's say i've got an impulse response taken at the right side 45 degress front, after convolution however the left ear sounds with almost the same amplitude (a bit diminished, though). It's true, the left ear is catching some of the right side burst, but it's not quite relating to it's REAL amplitude.
I attach an image of the patch. The fft convolution is based on the Johaness book patch.
Does anyone has any comments on this?