Pretty cool "audio Illusion" ... - Page 3

it's the same with me.

I dont know if this is binaural but its pretty damn awesome

Beardyman presents DolbyMan !

http://www.youtube.com/watch?v=BsF3n-_dZCA

Does anyone know why the frontal imaging is lacking in these you tube videos - are they working on solutions - what's happening with this tech?  I really wish for more support with this technology.  Its amazing.  It should be available with every movie track....why is this happening?

Quote:

Originally Posted by SanJoseCanJunkie 

So when you guys listen to these binaural demos, do you hear things go in front of you?

 

I only hear things go up, down, side to side, and behind, and when they are far behind I get a powerful, and annoying sensation in my upper spinal column.  This is on every headphone I've ever tried, and on every binaural demo that I've ever heard.

 

It's weird, because in some of them they actually say that they are in front of me, but to my ears they are behind me, and when they do circles around the head it sounds like they are doing semi-circles around back.  I've even tried turning away from my computer thinking that my mind is making it impossible to consider something in front of me because I know a computer is there, but that doesn't work either.  Seems odd since I don't have problems locating sound directions in real life.

 

Any others have this?  I know one guy on head-fi has the same experience.

Quote:

Originally Posted by SP Wild 

 

I find the same thing, rear imaging is superb - left, right, up, and down is eerily pinpoint at the rear - but severely lacking in the front.  What gives?

Quote:

Originally Posted by SP Wild 

Does anyone know why the frontal imaging is lacking in these you tube videos - are they working on solutions - what's happening with this tech?  I really wish for more support with this technology.  Its amazing.  It should be available with every movie track....why is this happening?

The most difficult thing to reproduce accurately in binaural recordings is frontal and rear sound cues. For the vast majority of people the left/right horizontal and vertical cues are easy to interpret but when it comes to frontal and rear imaging, it seems they tend to either blend together or simply not appear at all.

This is due to peoples own individual ears. The angles of the ears as well as the size and depth of the fossa and concha (and the distance from the meatus) are critical to how we all learn to localize and interpret sounds. Not surprisingly, the forward sound cues also contain the most complex and unique frequency spectrum when compared to other localized sounds. Even with the most rudimentary dummy head, achieving sound cues from all directions, except for the frontal cues, are relatively easy to achieve. It's seems the frontal/rear cues are always the most difficult to reproduce in binaural recordings.

One solution is to make a cast of your own ears and use them as the dummy's ears. The problem is that this will only work accurately for the person whose ear casts were used. If you use a set of simplified geometrical ears then it becomes more universally compatible but the frontal/rear imaging will also suffer as well as certain vertical cues. If you use averaged ears, you run into very much the same problems as before - non-universal compatibility being the biggest and frontal/rear cue problems being the next. If you eliminate the ears, then you don't really have a binaural recording as we know it and you lose total frontal/rear cues in most circumstances for a large majority of people. The type of microphone used is also critical as most cues in space are differentiated by the small differences in treble response our ears perceive. So a colored microphone is usually a no-no unless you compensate for it with EQ in post-production. In short, there is no perfect way to do a 100% accurate binaural recording for 100% of the population.

I have spent almost 3 years building different microphones and dummy heads for recording sound and the most accurate recordings are produced with the dummy head I last built. Virtually everyone who has heard the recordings it has made can distinguish the vertical and horizontal cues accurately. Only one person, afaik, has difficulty with the frontal/rear imaging....me. I have a few ideas on how to remedy this but I have yet to implement them. The funny thing about this is that on most recordings, I can easily distinguish between front and rear sounds but on others, the front and rear sounds become fused. Only time and many trials will tell if I solve this problem. Visit my blog if you want to listen to recordings done with my last dummy head project.

Hope that explanation helps!

NOTE: In the "virtual haircut" clip - there is nothing going on in the front, aside from ambiance, as far as I can tell. Everything happens from the sides and rear.

Thanks for the info. Very interesting!

There is some frontal sound with the matches audio clip, no?

Quote:

Originally Posted by Ultrainferno 

Thanks for the info. Very interesting!

There is some frontal sound with the matches audio clip, no?

With the matches recording there is frontal sound during the shaking part as he goes around the dummy two times.

some interesting discussion found here:

http://www.freesound.org/forum/viewtopic.php?t=13651