We've sent a verification link by email
Didn't receive the email? Check your Spam folder, it may have been caught by a filter. If you still don't see it, you can resend the verification email.
Started February 10th, 2011 · 46 replies · Latest reply by dwareing 11 years, 6 months ago
Recently I began to explore again the topic of "holophonic" sound, which is a (trademarked) term for very realistic audio that goes beyond headphones. On internet there is a lot of (unpleasant) rumor about it and some discussions (mostly unrelated to the technique itself), but nothing special. Listening to some demos however - I found, that there is a major difference between classical binaural (dummy head) recording, my approach and way the mr Zucarelli did.
While my experiments go beyond headphones because of some tricks that occur in naturally recorded audio (and can be used, without need of reverbs and delays), binaural sounds are very "arround" if properly recorded, holophonic effect is... interesting in different way. Vertical axis. If you ever heard the "matchbox shaker", you have probably noticed, that this sound - in a controlled way - goes down very deeply, even to the bottom of the spine and deeper. It is interesting to hear sounds that are placed so close to your backside and so far away from your head. Also - the holophonic effect is present (less, but it is) even if you listen the sound in 1 earphone only, the matchbox shakes up and down. Binaural recordings are "binaural" because the phase shift between left and right is strongly involved in order to produce 3D effect.
I have some suspicions how to achieve that "vertical" effect, but testing the idea requires some time, and DIY hardware. Meanwhile - here is my question and challenge. Were you able to create so deep vertical axis movement in your spatial recordings? Show us the results (-;
p.s.: here are the "matchbox shaker" examples (mono, stereo and holophonic; the last one you can try using only 1 earphone): http://planetaziemia.net/temp/matchbox.zip
Very interesting examples, listening on headphones really does give an impression of a vertical axis. What methods are employed in the recording stage as compared to binaural?
These samples are not made by me. I found them on the internet, and uploaded to my website for the reference. I guess these recordings were made by mr Zucarelli (the guy who invented and trademarked holophonics).
I'm not sure yet in which way this method differs from binaural recording, but I'm begnning to suspect "how". Yesterday I found some video interview with Zucarelli, and he refers this to holography and mixing the original sound with a reference sound. Describing the reference - he talks about "ringing sound produced by human ear". What I guess - this is about the asymetric resonant filter, that provides the filtered air as the reference.
From what I understand, dummy heads and binaural recording - they use microphones placed within the "ears".
I guess - the mic should be placed not within the ear canal, but behind the ear canal. Thus - the ear canal becomes a funnel with small spot (the ringing), that delivers filtered air as the reference. Ear funnel is asymmetric and soft/furry inside (like acoustic walls, which remove the reverbs), and "elevation" is produced by the brain as an "understanding" of this asymmetry. So, having the elevation reference, the rest of sound is registered in more open way (dummy head or simple field recorder with some absorber/reflector), as a normal sound.
When I have some time - I will do some experiments with ordinary kitchen funnels (or just made from paper) in order to check - if adding such spot - will change the colour of sound in the way I think. This funnel reference should be able to produce sound changes audible enough, to make the sound "spatial" in vertical axis even in mono recording.
"Matchbox Shaker" is a very good setting to make experiments, trying to replicate the vertical effect.
By the way - on my OMN playlist:
http://www.overtone.cc/profile/ayamahambho
you will find some tracks and excerpts, that are "beyond headphones", and also - some of them sound stereo/spatial, even if they are mono mixed and strongly compressed.
*
Another thing that comes to my mind. I heard about people who can literally see with their ears. Recognize shapes and textures around them (objects like stones, pilars, buildings, kerbs, and other things that interfere with air). With a little bit of training - I can do the same (perhaps not in so advanced way, but it is facinating, that you have a kinesthetic picture made of soundspace). There is something related to a resonant chamber behind the ear shell, and perhaps could be used to amplify the effect. Ear "membrane" is differently shaped than microphone flat membrane, so some adjustments should be probably made.
Just a thought. Try to eliminate, step by step - everything that you believe, that has to do with spatial localization. Flat your ears, close your ears with fingers, use some soft material to cover parts of your head, and so on. I think, that eventually you will discover, that the localization effect (especially the vertical axis) - is related to something that is within your head, and not outside. Then I would seek for - how the inner ear is constructed, and probably there is the answer, some spiral-shaped, well known and recognized structure. So instead dummy head you will get some snail/cochlea like pattern, that should cover the mic.
You can place a sound anywhere in the sound-field (left-right/in front-behind/up-down)using HRTF (head-related transfer function).I've found these impulse files work very well but so far have only been able to do it working on each placement seperately,would be great to be able to automate this function and just move around seamlessly in the sound-field,like the matchbox recording linked above......Interesting
From what I understood - HRTF is a rather dead direction in holophonic(s), and it is one of the "hot spots" of many THEORETICAL discussions on the internet. I don't care what is the mathematics behind the "vertical axis" (-;
Friends - ONLY sonic examples of "vertical axis" (and achieved depth) will speak for themselves. We should try to avoid "scientific theories", and focus on facts that we can test and measure. Theory will make feel you better, but only facts will make you hear better. We could spend months on talking who is right. Very quiet and silent months, with no sound around... (-;
p.s.: from my experience. usually - discovery is the first; theories can be then selected and adjusted to fit the discoveries.
ayamahambho
From what I understood - HRTF is a rather dead direction in holophonic(s), and it is one of the "hot spots" of many THEORETICAL discussions on the internet. I don't care what is the mathematics behind the "vertical axis" (-;Friends - ONLY sonic examples of "vertical axis" (and achieved depth) will speak for themselves. We should try to avoid "scientific theories", and focus on facts that we can test and measure. Theory will make feel you better, but only facts will make you hear better. We could spend months on talking who is right. Very quiet and silent months, with no sound around... (-;
p.s.: from my experience. usually - discovery is the first; theories can be then selected and adjusted to fit the discoveries.
When I was at uni I studied a bit of psychoacoustics... we briefly touched on trying to achieve vertical spatialisation through post-processing; IIRC the thing we tried as a group in the studio was EQing a recording to approximate differing HRTF... the result wasn't all that convincing as far as I was concerned (certainly not compared to that matchbox recording). In particular I felt that the source material we used (an aeroplane flying over head) was too obviously loaded with psychological semantics that would lead the listener to expect the sound to come from above...
It always struck me as a shame that there wasn't more of an overlap between 'pro-audio' type tech and gaming. Gaming soundcards do hardware accelerated HRTF, but not in a way that would fit in brilliantly with most music or sound design workflows AFAIK (although it could probably be made to, with caveats). I don't know how convincing those implementations tend to sound.
I've read in the past about HRTF implementation in SuperCollider that which I might possibly try to use with the rest of my SC system at some point when I get 'round to it... unless I go a different route and start doing more OpenAL style audio programming... I digress.
As for theory vs discovery... I believe that it is wrong to presuppose one will always precede the other; discovery and invention takes an open mind, the ability to recognise the new and put it into perspective, not be blinded by what you *think* you already know. That means being open to both tangible experience and also theory, but certainly not dogmatic about theory.
The problem with "theories" is following. If people begin, they can't just stop. Some time ago - I had a large conversation board under my wings, so I know how the dynamics of such discussions evolve...
...While no sound examples of working design appear. That why I propose to take some discipline, otherwise we sink in creative theorizing. I have nothing against theories, I build them too, everyday, to cover my discoveries...
*
I have mixed feelings regarding software-made sound spatial location. While plugins like Wavearts Panorama - probably work great (if you heard their "helicopter" example) - most of sonic material is not reverb-free. Therefore - when you try to mix field recordings with 3D plugins - you get yet another reverb, but not the real holosonic and controlled effect. At least - I don't know how to do it. Therefore - it would be good to create a physical design (using observation, imitation and elimination process), and then - glue to it some theory.
On the other hand - if sound localization is so well perceivable even in low quality sound files (freq limits, compressed, mono) - the spatialization (I mean this vertical part) is a result of a lot of perceivable processing. People on the internet argue if this is HRTF or not, but nobody was able to apply HRTF in a way, that would produce deep vertical axis. Or?
I would like to believe, that next person writing to this subject - will present their sonic results. When I finish my current ongoing project (and have some conditions and time) - I probably will try to do some tests, but I'm better in theorizing than building physical stuff (-;
This might also be of interest, 3D Audio at Princeton University
http://www.princeton.edu/3D3A/
I could not re-find, but I finally found them on vimeo:
http://www.vimeo.com/2856743
http://www.vimeo.com/2856748
Interview with Hugo Zuccarelli.
If you listen carefully, "as it is" - you will notice, that he is sharing (almost) everything vital about his invention and design. He is using "paralel/substitute description" (context related description; I don't know what is the english word for this; except "NLP" and "marketing" of course...), to hide details and show them at the same time. I do similar descriptive "cover up" from time to time, so I can recognize such word manipulation and what is behind. We call this hierarchic levels of abstraction; you go from more detailed view into more general view, and then go back to details, but in another context/view (and with another knowledge); vital cues remain in general the same, because they create the "foundation of truth".
Thanks for the link ayamahambho...interesting interview...not sure that he's not another potty professor remixing science to secure a patent but that matchbox recording and it's vertical axis is extraordinary....i've started making a few experiments myself using binaural mics but so far getting nowhere near the vertical axis that he achieves...i wonder if recording in an anechoic chamber helps the effect by giving better separation to the two channels?
....Anyway just want to say thanks for posting this topic...exciting stuff
.not sure that he's not another potty professor remixing science to secure a patent
I couldn't find the patent.
but that matchbox recording and it's vertical axis is extraordinary....i've started making a few experiments myself using binaural mics but so far getting nowhere near the vertical axis that he achieves..
Me as well, with poor results, it is intriguing isn't it.
I couldn't find the patent
...well he's got a patent on some holophonic speakers
http://www.acousticintegrity.com/acousticintegrity/Zuccarelli_Loudspeakers.html
interesting reference to phase balancing in this link....i was talking with a sound engineer today(35 years experience) who's never heard of holophonics but his first response after a brief explanation was that phasing problems will be almost inevitable in binaural recording and that this will have a great influence on the placement effect.Most of what he said went over my head..... but i gave him the link to the matchbox recording and i'm interested to hear what he thinks when i see him again in a couple of weeks.
Patents and Trademarks.
Trademark (TM) or Registered (R) is used to reserve a name. Patent is something, that does not need to have a registered name. It's a solution, design or something like that - something that can't be used in simiar ways by others without permission. With one exception of course. Natural phenomenons can not be patented. An example. Binaural beats as a difference between two signals - are natural phenomenons. Can not be patented. But what can be patented is a method of delivery of certain psycho-physiological states. A complex set of frequencies with a model of delivery, and the resulting effect. And you get the patent number. Method that uses binaural beats to alter states of consciousness, developed by Robert Monroe is an example of it. What you can do to protect your patents is to create a recognizable name and tratemark (TM) it or register (R). Method developed by Robert Monroe is called Hemi-Sync (R).
What Zuccarelli did - was to trademark an uncommon name. "Holophonics". What is behind holophonics - is probably not "patent" itself (if you patent something, then you share some of your details, which might be vital), but a market description of a phenomenon. The design that replicates such phenomenon could be patented, but patents are not good from commercial piont of view, not in this case. Similar designs could be developed with no restrictions, because they are based on something physical that we all humans have. You can patent the way you replicate the nature, but not the subject of your replication.
My proposal regarding development of recording techniques that will allow to replicate vertical axis in spatial localization - is to make it OpenSource.
Thanks for explaining.......this is a world of learning
My proposal regarding development of recording techniques that will allow to replicate vertical axis in spatial localization - is to make it OpenSource.
http://soundcloud.com/stus_binaural_space/binaural-explination-for-project-headphones-required
there is a nice vertical part inside.
I also share some recent/ongoing thoughts here:
http://www.freesound.org/forum/viewtopic.php?p=38287
General experiment with lowpass filter (hight order, sharp edge) and "matchbox shaker" example seems to show following. If the high frequency limit is c.a. 7kHz - sound is rather flat, not spatial, and has no vertical axis. If the high freq limit is c.a. 8kHz - the sound performs well vertically. Conclusion - resonant and/or filtering part of our head responsible for spatial hearing - has to do with freqs from a little bit more than 7kHz, and range between 7 and 8 kHz is enough to add spatialization. Bones? inner Funnels? Other tissue? We will see
(-;
You can do an interesting experiment. Take two glasses and sway with them around your ears. If the bottoms are pointed down, and glasses partially cover your ears from the bottom, the floor and trousers whisper sounds pretty funny. Glasses behave as a resonant/filtering chambers and affect the perception of sound very deeply. And now - imagine, that in your head - there are probably 1-3 small elements responsible for such amount of sound modification.
*
A thought, why the vertical axis is important.
Most people who do the binaural recordings - in my opinion do a common mistake, but they to this because of natural habits. Should the dummy head be standing or lying down? Think about it. When you listen to a relaxing backgrounds - you usually lie down. So what was above is in front of you, and what was in front - is now bottom, and what was behind - is above. If you lie down and listen to "standing head" - there is a sensorial conflict in your perception. But most people - record binaurally - with standing head.
Another example. Is it better to record the sound source in front of you or behind you? From what I hear - what is in front - is louder, but less spatial. What is behind - is more spatial. When you hear a sound in the night, and try to find it - are you pointing your head "in front", or rather you use a part of "behind your ear"? "Listener experience" seems to be a little different from "visual expectation", but we are used to visual expectations.
Do some experiments and tell about your results.
(-;
@Microscopia - I see you are repeating "known" theories of someone else, that it is assumed that everyone knows. I think, I've viewed that one too, some days ago. I agree - this article brings some information, but we still live in 1g gravity. Besides in air, when you are in the glide/sailplane for example - sound localization is the least of your concerns....... (-;
What can I say to that article. Not everything is taken under consideration in science, which rather prefers "sterility" and "singularity" of measured events. Therefore I probably partially agree and partially disagree with. But I can add, that I personally know (and heard about lot more) scientists, who must (literally - must) limit their research and theories to so called "mainstream", because if they don't - they are excluded from scientific society and have difficulties to find better job. Sure it changed in last years, but this trend is still present (try to google in videos - a lecture of Dean Radin about spooky science). Mainstream means - control over the flow of information.
BTW, while "Localization of Sound in Micro-Gravity" article has a lot of words (including a sentence in Conclusion section: "The data presented in this paper are far from conclusive." - did you found there sonic examples or practical cues how to record spatial sound? Well - I did not.
"we humans have not evolved for vertical sound localization" - that's an example of "unaware/involuntary mimicry", Susan Blackmore would call it "meme theory". Someone said something and certainly this is truth, because he/she was PhD (-: Most of us were taught "what to think", not "how to think". I often begin my day with an affirmation created by Bob Monroe: "I am more than my physical body..." Can you believe, that I'm able to imagine movements in multidimensional spatial space or through time? Did we evolved to that? Who said "not"? Beliefs based on other beliefs. If my experience limits my reality to "knowns", how can I experience something new, if I expect to experience what I know/believe? The art of listening means - sound on one hand, and learned ways of perceiving on the other hand. I don't think that this is a matter of evolution; we rather have a big potential to manifest, but we still use a small percentage of it.
Okay, back to practice.
"Matchbox shaker", when it's down - I hear it around/behind my ass (anatomic), no matter whether I stand or lie down, sure. Well, perhaps not so deep down when my ears are blocked during flu or cold or something like that. But what I mean - is a practical recording cue. I usually listen to nature recordings in lying position, not sitting straight. I often fall asleep with them. But these landscapes are usually recorded from sitting straight position. If I lie down, then my sense of touch and gravity has different reference, and my hearing has a different reference, which is confusing, because sky with rain and thunders - begins to appear to below me (90 degree sky shift). You heard the "Storm 4D" (thanks again) I made recently (-;
Have a nice evening!
p.s.: here is some discussion about binaurals:
http://www.gearslutz.com/board/remote-possibilities-acoustic-music-location-recording/493726-holophonic-vs-binaural.html
http://www.gearslutz.com/board/remote-possibilities-acoustic-music-location-recording/489576-best-bets-binaural.html