Does anyone know how binaural microphones actually work? I know the super expensive ones use a special, calibrated, EQ-ed-like dummy head, but there are a handful of binaural microphones that don't have that.
3Dio is an example. I have a pair of Sound Professional Master Series binaural microphones that I use in some of my YouTube videos.
Here's a random street performance I captured with those microphones if anyone is interested (I was using a Tascam XJ2 as the preamp since these are electret microphones): https://dl.dropboxusercontent.com/u/2816447/2013-08-14_-_UVillage_Concert.flac
Yes. The technology is pretty well understood, but I'm not sure where to steer you as there is a lot of ground to cover. However, start by looking into the HRTF - everything else will sort of tie into that once you have a good grasp upon that concept.
I can't remember (probably gearslutz) where I wrote about the different types of mannequin heads, and the various types of equalization that they employ, but again, there are many, many ways to do this...with varying degress of realism, but if I find the thread, I'll edit this post and add it.
ADDED: Here's a link to one thread in which the issues surrounding mannequin head microphone types and types of equalization is addressed. My comments on the thread are really geared to the differences in binaural equalization (as well as why the 3Dio mic isn't really a binaural microphone), but this might be a good thread to check out (I've written a lot more elsewhere on gs on the subject of binaural (lots), so if you're interested, you can look through my posts (my handle is "Mark A. Jay" on gearslutz - at least for now anyway) and related posts by others on the subject). Anyway...back to the link:
In general, it's not really a true binaural recording unless a mannequin head microphone is involved - in principle. There are other mics out there that use a sphere with mics on either side that achieve much of a true HRTF-based recording (because they mimic - to some degree - the shadowing that a true mannequin head would, and this makes a big difference than if the mics had simply been separated by the same distance (as in the head) without the obstruction between them). The presence of the 'obstruction' (and its geometry) that is the head portion is really, really important to spatial cues in a binaural recording...but there are several factors that govern this (the realism), and this is way beyond the scope of this post.
Another place to learn more about this stuff is the Perceptual Audio Forum (I'm the Group's Moderator) on Linked In, so if you or anyone else is interested in joining, the posts there may be of interest because the discussions tend to be pretty heavy on binaural as well as signal processing techniques designed to emulate binaural with speakers (i.e. cross-talk cancellation schemes). Here's the link: http://www.linkedin.com/groups/Perceptual-Audio-3722860/about
Anyway, if you're interested, just send a request to join the Group.
There are lots of resources out there about binaural and the technology surrounding it, but as I pointed out elsewhere (possibly in a different thread), the monograph "Binaural Technology" by Mme Rozenn Nicol (of Orange Labs) is a really invaluable text to have:
I hope this help. I'll have to check out your link sometime...busy day today.
Edited by immersifi - 5/15/14 at 10:38am