This is a bit off topic but I didn't want to start a new thread. We've discussed binaural recordings here, and the crosstalk-cancellation needed to play them properly over speakers.
There are a number of commercial DSP units that do that to various success, but are generally very expensive. The only software that I've found is this, and it's not free:
http://www.wavearts.com/WaveSurroundPro.html
The method is described in Gardner's thesis, along with making binaural signals from non-binaural ones.
http://sound.media.mit.edu/Papers/gardner_thesis.pdf
Detailed HRTF measurements are available freely, so I don't know why I can't find any free software that will do this. It would be great to find a tool that would convolve a music file with the appropriate filters, so I could rip, process, and burn all my CDs for a better speaker listening experience.
Of course, I could just put a physical barrier between the speakers, but people would laugh when they see it.