3D Printing Binaural Microphones

Binaural audio is probably the coolest thing you can listen to with a pair of headphones. Instead of just a single microphone, binaural recordings use two microphones, set inside an analog for a human head, to replicate exactly what you would hear if you were there.

The only way to record binaural audio is with fake plastic ears attached to a dummy head. Most of the famous microphone manufacturers have something like this, but with a 3D printer, anything is possible. [Carlos] created his own binaural microphone using a 3D printer and went through the trouble of creating a few audio demos. The results are weird, like [Carlos] is whispering into your ear.

The ears used in this microphone setup are taken from a Thingiverse project by [Jonathan March]. This model did not properly model the ear canal,and didn’t have any way to mask the sound from ear to ear; this is why the professional models also include a head. [Carlos] fixed these shortcomings and created a few 3D models that accurately model the human ear and head.

There’s also a simple stereo microphone amplifier for this project that is designed to fit right between the ears. This amplifier was designed in KiCAD, and the PCB is single sided. It’s not quite simple enough to assemble on a piece of stripboard, but [Carlos] did manage to manufacture it on some copper clad board in his mill.

The results? It sounds awesome. [Carlos] put together a demo of his microphone, link below, and it only works if you’re wearing headphones.

62 thoughts on “3D Printing Binaural Microphones

        1. Thanks David!! There is actually a whole section of YouTube dedicated to those kind of “tingles”, you can search for “binaural ASMR” (though it can get pretty random very quickly lol)

    1. I’ve done that a few years ago (does anyone remember Mini Disc recorders? :D ). The effect is stunning, if you playback the recording via headphones. But it is also true that you can hear all kinds of strange body noises ;-) But still worth a try, if you like unconventional ambient sound recording.

    2. enter Soundman Microphones. they look like in ear headphones, but the signal path is reversed. they come in a nice wooden box, including a preamp and they have several versions from very sensitive to near deaf. with the latter you can literally stand in front of a pa system and have no distortion (but your own ears wil suffer..) i have the classical studio variant and they are amazing. you have to learn how to wear them, but taking a ride on a metro or bicycle is no problem.

  1. For music or sound recordings, I would rather record outside of the ear. Since when I play it, my ears/head will distort the sound accordingly. I do not want the sound to be distorted once by fakes ears and then again by my own ears.

      1. And you get the same effect with two microphones placed an appropriate distance and orientation from each other with an acoustic barrier in between them. The fake ears don’t help.

        1. A Jecklin disk setup can’t distinguish sounds from above or below (there is nothing for the sound to be induced with HRTF from the folds and parabola of a pinna) therefore are just stereo. If this chap would have demonstrated UP and DOWN – it could have been a far better example. However, as with ALL ears, they are all different in shape and every single person who listens to the recordings will hear them differently because of that. Once someone realises what the pinna actually does to the sound, they would not make any assumptions (like most people do) at all.

    1. I believe the ears physiology evolved to filter sound which isn’t inherently distorting it from the perspective of electrical amplifiers which we design not just to boost a signal but also to filter. Distortion to me refers to when the input signal is too large so the output saturates as it is just outputting the power source voltage… or distortion from power amplifiers (crossover distortion) or when you have unwanted frequencies.

      TLDR; The ear does an amazing job of filtering unwanted frequencies… just like the electrical amplifiers we make

      I would use the acetone smoothing method and then cast the ears in prosthetic skin (silicone)

      or skip the 3d printer and just make silicone molds of peoples ears… everyones ears will filter uniquely

  2. Whispering! I won’t print any f words. Just about the most disgusting sound that humans make. SPEAK UP! Worse than farting.
    I love binaural and use a commercial head with updated condenser mics. I moved the mics up to the entrance of the canal, the old ones were at the end of 2 inches of curving tube to face one inch big dynamic mics on handles shoved up the neck.

    The reason that this is not used in studio recordings is that studios rarely make live music so it cannot be captured live. They have no acoustics, just deadening. Mics are shoved into every instrument and mouth, so the whole sound never happens. Plus, with binaural no engineer is needed. No knobs to twiddle. No altered sound.

    The sound of a live session in a club where everyone is playing together hot is a real treat in binaural. Sadly almost no recordings are made this way. I point at the knob twiddlers and sound sanitizers, this stuff has been going on since Les Paul. Binaural since 1881.
    This weekend was the Indiana Fiddlers Gathering, I didn’t go. I have many hours on cassette of stage sound (no monitors) and the best campground pickn’ that went on till dawn. Yeah-Haw!

  3. A few model inaccuracies I think would help. Material around the setup to emulate the basic headshape, like a $1 ball with holes in the sides. A coating of latex like material inside the ear canals to soften the noise as it bounces along, I think the plastic will cause higher frequencies and pops to pick-up more readily than they should.

  4. Is it me or has Hack a day and some of its commenters become a bit snooty. Why are people criticizing the guys choice to build such a project. If binaural recording is not your thing, go work on and comment on the projects you do care about. This site USED to be about the hacker/maker/diy community back in the Caleb Kraft days, but seems to have drifted into something else. I finally find a post that is interesting to me and a bunch of haters are pummeling the poor guy. People kinda suck.

    1. Yes, I have become more snooty since reading HaD!
      Many commenters have increased my insight to the featured projects.
      This makes me a more discriminating reader.
      As the old saying goes, “with great power, comes great responsibility”, so I now MUST share my enlightened thoughts with other HaD readers, especially any new ones that have not ascended to such heights of enlightenment.
      It is in the very nature of Open Source, for if the source is not open to feedback, it is surely not Open!

    1. That is a very interesting thought indeed. If you have multiple mics it is actually possible to merge their signals digitally to achieve “beam forming”. These guys achieved it recently and give a nice explanation: https://www.youtube.com/watch?v=U88dTxcpf6c

      Listening as other animals would be lovely too! Cats have higher gain ears which can move independently. We should just find a 3D model and try to fit them with mics! :D

      1. That would be great if your cat could wear headphones.
        I thought about this very idea too, but then realised as we are not a cat, dog, bat, fox or elephant, we wouldn’t know what we were listening to, as the HRTF information would be alien, yes we would hear what they heard, but we wouldn’t make any sense from it, like we do if we hear something rustling in a bush for example. With their pinna having such small parabola and other folds, I can only assume that their hearing has a more “pin pointing” ability to focus onto a particular direction of sound source. Understanding how our own pinna work is a fundamental which has to be realised before any assumptions of other species can be speculated.
        The folds of the pinna are VITAL for localization or perception of direction.
        This is a recording of a fixed position microphone recording of a simple knock which is then simply speeded up – Do not move your head during play back.

        Thus proving that the pinna folds and parabola play a vital role mostly over looked by the most blinkered, oops, I mean most educated/researched types of persons.

        :P

        1. I’d like to try listen to sounds passed through animal’s HRTF just out of curiosity, to see whether there are any obvious similarities.

          That sped-up knocking sound explains the matter very well, thanks for the link!!

          BTW congratulations on your youtube channel “Frank the binaural microphone”, your videos are great! ^_^
          And the idea of having a live streaming from your garden in binaural audio is very nice, I had never seen that before. It must have been a challenge to make the mic waterproof.

          1. Hey carl, thanks for the flowers, the YouTube channel has only just started trying that ASMR stuff, I’m aware of what it is, and one aspect of the experience, however, it was a shock that I seem to have what it takes? I was actually trying to “show off” the microphone to that community, and suggest that these mics can be used for many other things too. I’ve gained nearly 1,000 subscribers in the first month of posting that ASMR video, took me by surprise indeed.

            I think the binaural microphone in the garden could very well be the only one of it’s kind streaming at present? it’s throwing it down here with rain as I type! Poor Frank! lol. The balaclava seems to keep the water out from the inside of the dummy head, there are plenty of places where water could get it to be honest, though the microphones are actually quite high on the inside of the pinna, so if water did get into the ears it would never be so much as to flood them. Perhaps I should do a video of that too.

            I’ve just updated a playlist, about inverted sight. Saw one of those many years ago as a kid, quite amazed that I found the same one on YouTube, but even more surprised that there are a few, and one lady could ride ther bike in no time at all!… which leads me on to the possibility that HRFT can be learned – that’s NEW HRTF so that you could (if used often enough) percieve directions from visual cues when wearing the new HRTF induces (different pinna). once removed, to be able to go back to your original ears without any real trouble at all. I think this proves the point? Who wants to try that experiment? Dog or horse anyone??? lol… Seriously though if this is the case, then it would mean that a HRTF STANDARD could be used (once learned of course – like riding a bike).

  5. I like the criticism of the people being so cynical in this thread … surely experimentation is a thing still to celebrate, if you want to be critical be constructive with it. I just love people making! I don’t have a community of like minded experimenters with technology where I live so seeing it on here and to be able to interact is a privilege, not something to denounce.

  6. I would think that if you built a system like this, set it up, and did a recording using some impulse responses, one should be able to extract a transfer function from the analog stream. After you have your transfer function, so long as you have 2 microphones in opposing directions, with the sound absorber in the middle (as a previous commenter mentioned), it should be possible to recreate the corrected sound without walking around with a pair of ears. Simply take the stereo stream and run it through your transfer function corrected post processing!

    1. It can be done in software – look up “head related transfer function” or hrtf. How ever it is not as simple as putting a stereo mix through a convolver as the stereo mix doesn’t contain enough information about the location of individual sounds. The best you can do with a stereo recording is to simulate 2 virtual speakers in a virtual room. The same is done for mixing down 5.1 surround sound for headphones, but 6 speakers are simulated. It is much more interesting when there are more sound sources, which can even be moving.

      1. Ok with pleasure, now I have one stereo microphone like one pro :-)
        Listen the result with headset, the sound is incredible !!!

        And, the bonus is, and I already said, the wind immunity is very good. :-)

        Pictures :





        For adapt to Gopro, you need one mini usb 10 pin connector (easy to find it on the web, carefull take a short one, you can find 2 models long and short ) and one resitor who will commute for external microphone ( you don’t need 1/8 watts, like on this picture, one cms is enough )

        And, because you will ask me for the steadycam ? You can find it here : http://www.feiyu-tech.com/G4-en.php :-)
        This one is expensive, but the result is : Incroyable, how we say in french ;-)

        Your credit card will be burn, (with the steadycam) but you will be able to make some realy pro vidéos.

        Enjoy…

        1. Yeah it looks incroyable!! Thanks a lot for the pictures :D
          Best part is that you don’t need the circuit as the GoPro has audio input. It is so simple like this!! ^__^
          A demo video would be amazing too :P
          Thanks again!!

    1. I’ve tried it, the HRTF isn’t effected that much unless the sound is a high db, then you can get the material resonating, which contaminates the separated sounds with mono sounds (which appear as sounds inside your head- like stereo does), as long as the microphones are insulated from eachother and the pinna will not resonate with sound then it should be perfect. My own 3D printed ears where hollow, so I filed them with Decorator’s Caulk. I’ve a live stream of my garden below, as the head is polystyrene too much of a loud sound will sound strange, I’ve taken it out on the streets (very brave) and made many recordings with it, I’ve even been to Elstree Studios in London with a few of my home made microphones :)

      http://locus.creacast.com:9001/liverpool_ormskirk

  7. Hi.

    Good idea. But also this “3dio free space” microphone are critizised.
    FYI, a thread from a froum:
    https://translate.google.com/translate?sl=de&tl=en&js=y&prev=_t&hl=de&ie=UTF-8&u=http%3A%2F%2Fwww.tonthemen.de%2Fviewtopic.php%3Ft%3D2237&edit-text=&act=url

    They criticize the missing head.
    But also missing shoulders.

    Are you interested in developing a full binaural dummy head?
    With optional shoulders :-).

    As a basis a real human massive skull.
    Flesh / Skin is fillable (lots of voids). Fillable with wax, plasticine or ballistic gel etc..
    Fillable brain space etc.
    Detachable ears to insert and hold microphone capsules deep in the artificial ear canal.
    Etc..
    A real human (male) head has ~9lbs.

      1. A dummy head? This is standard in binaural microphones.
        This “ears on a box” is the new “Light version”.
        What exactly does not work?

        I mean a complete head including skull in it.
        A single part.
        This meat etc. above the skull is firmly connected with it.
        But fill up with large cavities.

        There is no reason that this does not work.

        You can also cut off the head from the nasty neighbors, but this is not legal, and after a short time begins to stink.

        1. Did you design the head?
          I design binaural microphones and saw something that most overlook through assumption.
          The error is clearly visible for those who comprehend the function in your first message.
          A reply asking why I came up with the conclusion would have been presented by the facts, but you didn’t do that.
          Have a nice day.

          1. No reply on my question (“why”).
            I am not a 3D-Designer. No knowledge. Is this easy?
            Are you upset? I was not wrong, it would be possible.
            Of course that would be a head in real size and possibly weight (9lbs), but it would be doable, and would work.

            For example, a professional dummy head microphone:

  8. I can’t see any context for this “lower-class adage”.
    Just an unnecessary aggressive comment.
    This was not a personal criticism, just a suggestion.
    Also “3dio free space” Is not for good quality binaural audio.
    It is a fact that the result of “small box with ears” is worse than a dummy head as it has been used by experts for over 40 years (~1972).

    Btw, your username is “binauralhead”…

    1. I think you were triggered.. and then some (I’ve no idea why, all those books perhaps?)

      Going by your comments on the 3Dio, you must be talking from one of those severed heads you mentioned before.

      Btw, your username is “binauralhead”…
      You’re kidding … ?? Right?

      And still, you have no idea what you are talking about, suggest take head out of book, and practice what you preach.

      For the final time, goodbye, and have a lovely day/evening/morning/whatever, I’ve switched off my ears.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s