[j4cbo]

With the return to popularity of multichannel audio, a lot of albums are being rereleased in surround-sound format. This is all well and good for people who have multi-speaker systems, but doesn't help us headphone types much at all.

It seems to me that it should be possible to use a head-related transfer function to mix a multi-channel source into a binaural track designed to be played back over headphones. If done properly, the three-dimensional soundstage would be entirely preserved. I would think this has been done before, but I've spent quite a while searching for implementations and demos and found nothing. True binaural recording is difficult and rare, but 5.1 mixes are becoming increasingly common; why not turn one into the other?

Anyone tried anything like this?

 

[werdwerdus]

You could get a set of binaural microphones and record your 5.1 speakers as they play the music

I think someone has most likely done that before.

 

[Uncle Erik]

It wouldn't work. Binaural applies to how the recording was made. You need to use a dummy head with simulated human ears, et al. to get the effect. The typical 5.1 (or 4, or 2) channel recording has microphones widely separated. Also, they typically mike each instrument and put them on separate tracks. If there is a mike on each guitar, the singer, the drums, etc., then you will not get the cohesive effect of binaural. There would be no way to create binaural from that mix, either. To get the effect, everything has to be "heard" by the recording device the same way you or I do. That's why you use a dummy head with microphones in its ears.

It's a shame, though. Binaural is the best there is for us headphone geeks. Hopefully, it'll catch on now that we live in an iPod world. Sooner or later, I'd like to rig up a dummy head and take it along to open concerts and the bar shows, where some of my friends play.