Teach yourself to echolocate (2018) | Hacker News Reader

Teach yourself to echolocate (2018)

atlasobscura.com

247 pointsAnon8475 comments

This is often used by audio mixing engineers and taught in a roundabout way at schools and studios. We think a lot about where thins “sit” in the mix. Proximity wise, and even height wise in a stereo mix. Eventually you learn how to locate things in headphones and it’s a really weird sensation when you realize you can do it. The kicker is we start out by simulating real environments in mixes, but then end up having to simulate what people expect from the medium as opposed to real life. For example something I learned doing video audio, if someone is writing something on a train, viewers expect to hear the pen on paper. But irl, there’s not a chance it’s audible. Explosions are always distorted because microphones end up clipping due to the volume, etc.

A great book on spatial simulation is The Art of Mixing by David Gibson. Older but forever relevant

> For example something I learned doing video audio, if someone is writing something on a train, viewers expect to hear the pen on paper

Just yesterday was watching Territory season 1 where the characters have an intense suspenseful, almost whispering "serious voice" conversation while standing next to a running helicopter, without even raising their voices which took me out of the scene.

So the question is, do viewers want it, or do know it all producers say people do and put it in?

When they say viewers want it, they mean just about 90% won't notice. Most people haven't been close enough to a running helicopter to understand.

I'm having problems watching movies at all, there is so many things breaking my immersion. :-)

> When they say viewers want it, they mean just about 90% won't notice.

Even more than that, they will notice if you don't it the "wrong" way that they've come to expect. This is called The Coconut Effect: https://tvtropes.org/pmwiki/pmwiki.php/Main/TheCoconutEffect

Ah, had no idea it had a name!

My clearest memory of that was me as a kid watching a Bond-movie where a sportscar makes a screaching sound when driving down a sandy beach. I turned of the TV and don't think I ever saw a full Bond movie after that.

The list on the page you linked had one thing that isn't toally correct though:

>The very specific (but entirely unrealistic) echoing thud that is heard when all the lights are turned on in a large spacenote .

That sound is realistic if it is an old building with the heavy type of power relays or whatever they are called. They do make that sort of sound if the acoustics are right. They could be set up with timers so they don't start the lights at exactly the same moment to prevent overloading the fuses.

Have you never walked through sand and heard it screech or squeak? It's definitely a thing.

https://en.m.wikipedia.org/wiki/Singing_sand

Surprisingly, it’s also present in live sports, see for example this article (in Finnish) about sounds in winter sports broadcasts: https://yle.fi/aihe/a/20-10005843

You went on quite many topics there. Could you expand on the proximity and height? Fascinating

The closest analogue I can think of is how due to practice now anyone can close their eyes and imagine typing entire essays how they know exactly where the keys are. Try it.

I have played alot of videogames & at some point identified, how can I guess, the source of sounds. Guess, because it's nowhere near actual approximation. Most often, source is guessed by context. E.g. The door knocking sound illusion, which was used to troll streamers.

Then you have directional localisation based on delay between ears, difference in volume & properties of reverberations. Things to the sides are going to arrive in either ear at different moment. Add source if first echo & you have confirmation that a sound is coming from either right or left. The more directly to the side is the sound, the bigger the delay between ears is, so you get approximate angle.

Now we consider sound muffling, caused by shape of our head & ears. Things in front are going to sound clearer in the opposite ear, than sounds from the back.

The same principle is used for detection of height. Things below are going to get muffled, things above will be clearer. In reality, feeling sounds with the whole body helps in source localisation, which can't be emulated with headphones.

Ive seen a BBC doc that tested this. They had people use putty to change the shape of thier ears, resulting in an inability to judge the height of a sound. Given how differently-shaped ears are, as opposed to the inner structures which are virtually identical, this result points towards a learned skill rather than something genetic. We each must learn how our paticular ear shapes modify sound.

I once lived in a shared 4 story house and I always intuitively knew where everyone in the house was, even if they weren't being loud. You could just tell where everyone was based on how the house creaked in response to footsteps. We had someone new move in once who walked very quietly and it made me feel slightly uneasy because they were sort of invisible to my hearing.

There is a video for the art of mixing. It is indeed fascinating.

https://m.youtube.com/watch?v=TEjOdqZFvhY

If you know how to touch type you should already know how to do this. I know for me it was a requirement in my 7th grade class.

That's not at all what echolocation is. What you describe is locating the source of sound using binaural hearing (similar to how we can gauge distances using stereoscopic vision).

Echolocation is finding out distance to objects (not sound sources!) by sending a sound wave in a direction, and listening for echos that bounce back. Hence echolocation.

The only sound source is you.

It's a form of active sensing: literally how a submarine sonar works (or radar, for that matter). Bats do it, too.

This has very little to do with "locating things in headphones", as that is entirely missing the active part in the first place.

Then, locating sound sources using binaural hearing is not the same as analyzing the scattered echoes when the sound source is you (relative to yourself, you know where you are already!).

It's interesting that this is currently the top comment. I wonder how many people read the article before engaging in this discussion.

> literally how a submarine sonar works

And dolphins and whales, no need to go to submarines.

Interestingly, it took until after the invention of SONAR for the theory that bats navigate by echolocation to be accepted. The theory that bats use hearing for spatial awareness was first proposed in the late 18th century, with experimental evidence, but was rejected by the scientific establishment for more than a century. People didn't know marine mammals used echolocation until the 1950s.

I didn't know this, but the intuition that a tech example will be easier to grasp than an example from biology was why I mentioned sonar before bats in the first place.

Fascinating to find out that the scientific community had this kind of bias as well.

yeah- yikes I did not expect that to happen. My comment was extremely tangential and there were a good few comments already when I added it. I completely agree - it’s not about echo location

This comment captures a lot of important detail about echolocation.

Thanks! I'm glad you found it useful.

I think what you mean is that all your examples don't work when recorded. But a human being in a train may hear the pen on the paper.

You can absolutely hear someone writing in a room.

Whether it's audible on a train, depends on how insulated the train is.

I get the OP's point, but indeed this probably wasn't the best example.

There is a wonderful book about the blind man who was probably one of the earliest innovators in using a cane for echolocation. "A Sense of the World" by Jason Roberts is the story of James Holman, who traveled the world in the early 19th century despite being blind, often being in a great deal of pain, and having limited mobility.

Wow it just struck me that the cane is for echolocation and not just obstacles. I did learn that the blind have keenly adjusted hearing.

Blind people can use a cane for echolocation, but it's not necessarily reliable (especially in a busy city), and to be honest I'm hard-pressed thinking of any blind people I know that actually know how to do that.

They are probably constantly aware of surfaces the cane is bouncing off & use it as a redundant confirmation of what they "see" by touch

I don't, that's for sure!

Wha? Well, there are blind people who barely know how to use their cane, but... Let me explain.

Manfred Spitzer once wrote that he thinks there are two groups of people on this planet who really have good audio location capabilities. Blind people and conductors. Conductors because they need to be able to listen to a particular performer, to isolate them from the rest of the orchester. And blind people, because we use the ear to navigate the world.

Now, I actually use everything around me as a source of sound. Tapping with the cane is one of them. However, if I want to "scan" my environment, I usually make a clicking noise with my tongue.

But those are the a small part of the game. The rest of the noises I use come from outside. Just a small example, before I loose myself in thsi comment: I can hear poles and trees on the sidewalk. Not because they emit so much sound, but because they eat it up. If a car drives behind the pole along the street, I can actually hear the point where the external sound doesn't reach me, infering that there must be a pole or a tree. Echo location is not always about what you send. Its m6ore about you learning how the sound waves around you behave. Sometimes, but this is getting borderline esoteric, I can hear the materials involved. Walking towards a wooden wall sounds destinctly different from walking towards a concrete wall...

Fascinating. That’s a lot like Passive Coherent Location (PCL), if normal echolocation is like radar.

For most it's not, echolocation is not widely taught.

You couldn't teach me to do that. Also I don't have super powers. I really do wish I could fly though. Give me wings any day!

I noticed when I was younger that I can "hear" when I'm near a wall, or just generally get a sense of the size of space I'm in solely via sound, but it never occurred to me to make a "ping" to sense the resulting sound reflections! Super interesting - I will have to try this. Probably my prior spatial sense from hearing was based on really subtle background noise of either my own footsteps or just environmental sounds reflecting around, I guess? I always felt like I am "hearing the room I'm in" but never quite knew how else to describe the sensation, but knowing people quite literally make a clicking noise to echolocate suddenly makes it much more clear!

Comb filtering (https://youtu.be/Amj4UevyRfU?si=5pwjHKwAw6bdmG2x) plays a role in this.

I was listening to a podcast and realized I could hear the speaker turning pages under the microphone by the way it affected their voice in the microphone rather than the rustle of the page. It was pretty wild. I could ‘see’ it before i recognized what was going on.

Thanks for the video, I wouldn’t have guessed it shows well on a video.

Kinda off topic but I’m on a brand new phone (not logged in and no history) and the next video suggested by YouTube is a French fascist promoting (actual) nazis policies. Why would YouTube do that?! It has absolutely zero connection with audio topics. I just have my OS language set to French. That’s so worrying for the youth with being exposed to pure hate for no reason.

Because it makes Google money.

At a guess, it attracts two audiences - people interested in the content, and people aghast at the content. Twice the pulling power?

I did that, too, but I took it a bit further, possibly as a product of not having vision for about a year as a toddler, other than "light" and "dark."

Here's an idea, use a (portable) ultrasound emitter device that sends a pulse every set second and use an earpiece receiver that produces equivalent acoustics in audible range. The setup may reduce "tongue-fatigue' and the ultrasound will travel farther and can reflect off smaller objects. Would be an interesting experiment if nothing else.

I suspect it would be a very challenging problem for the earpiece part to recreate the directional info as well as our own ears, since the brain's ability to detect the direction of sounds depends on the shape of the ear itself.

Maybe a simple mechanical clicker device like those used for dog training could be a useful tool.

I have one of these - purpose built for blind people to use. Pretty sure I picked it up at a NFB convention in Atlanta around the turn of the century. (NFB = National Federation of the blind)

How well does it work?

I will get back to you on that. I just found it in my desk. I myself am sighted, and have never used it for echo location - yet. Looking at it I realized that it was given out by my old employer: Blazie engineering. Coincidentally, I just ran into the founder recently. Next time I see him, I will ask him about the feedback he got on these clickers. I will say that after twenty years in a drawer it sounds good to my uneducated ears.

please ask them if this or any others were commercially available. Thanks!

If the wavelengths are not far from audible, wouldn't it be mostly a translation?

Seems like a fun use case for spatial audio on the airpods.

See here: https://www.youtube.com/watch?v=PD3Y1l8XyUw

Another route would be to mix the ultrasound with another sound closer to the ear, then there is no need for an electronic ear at any point. The interference between sound can cause the inaudible frequencies to become audible.

Just on this topic, would it be possible to make a whistle to do the same thing? I.e. crafted so it emits both ultrasound, and the audible counterpart which interferes with it to make the return audible?

Perhaps it could be such that the ultrasound warbles whilst interfering sound does not (or vice versa), which would make the sources easier to distinguish also.

relevant video: https://www.youtube.com/watch?v=PD3Y1l8XyUw

It appears that the hardest part of echolocation for humans is the "produce a directed, crisp click" part. The "process the sound" part is readily handled by our brains with a relatively mild learning curve.

A few years ago, I was a frequent visitor to a public swimming pool, same day, same time. Apparently with similar swimming habits, there was an elderly person who was humming all the time. In the pool, in the shower, in the lobby, rather loud but not too much. I initially attributed it to elderly quirkiness. It was only after several encounters that I realized he was blind. The point is, I only figured that out when I saw him with a white cane outside, not at all by the way he moved inside or used the objects - he was navigating the space just like anyone else, and it was a rather crowded place. That was the day I learned that echolocation in humans is a thing.

I feel like a pool is an especially good place for it too. Everything is hard surfaces, providing a good echo. The ambient noise probably helps a lot too with not needing to make noise yourself.

I don't have much sense of the reasonably expected or maximium performance of this ability. They give a few sensational demonstrations, but how accurate is it (as in complete, correct, and consistent)? How fast? In what environments? How granular? Is it practical for everyday use?

For some interesting context, here is a description of dolphin echolocation:

https://www.britannica.com/animal/cetacean/

"The amount of information obtained by an echolocating dolphin is similar to that obtained with the eyes of a sighted human. ...

Toothed whales use extremely high frequencies, on the order of 150 kilohertz, for refining spatial resolution from their echoes. They are capable of “seeing” into and through most soft objects such as other dolphins, though the effectiveness of toothed whale echolocation drops off at distances greater than about 100 metres."

Maybe with enough practice ...

Sounds travels very differently in terms of velocity scatter etc in water though. If somehow dolphins were capable, they would not nearly have the same echolocation capability on land

E- chocolate

Don’t get it.

Oh..!

Yep same. "Teach yourself to eat chocolate."

had a french voice in my head going "teache yourself to eatee chocolate"

Sounds good to me!

LOL, I passed by the article a few times reading the same thing. When I finally clicked in thinking it was something to do with the windows package manager. Something about the capital E. As soon as it was spelled with the little e, it clicked.

- Humans Can Learn to Echolocate (Livescience, 2015) https://news.ycombinator.com/item?id=10699105

- How humans echolocate 'like bats' (BBC, 2018) https://news.ycombinator.com/item?id=16782557

- Humans Can Learn How to 'Echolocate' in 10 Weeks, Experiment Shows (Sciencealert, 2021) https://news.ycombinator.com/item?id=27404132

- Teach yourself to echolocate - 106 comments https://news.ycombinator.com/item?id=18208334

I initially misread it as “Treat yourself to a chocolate”

I misread as "Teach yourself to e-chocolate" and was even more confused.

Maybe only distantly related to this, but thought it worth sharing that when I visited Seattle for the first time this year I caught a show by a band called La Cerca at Central Saloon and loved their song “Echolocation”, with ethereal sounding guitars, including the bass: https://www.youtube.com/watch?v=1NYeqA2Kve8

Meanwhile, after reading this article, I have my face one inch from a wall making clicks to see if I can hear the echo. Hopefully, nobody sees, or hears, me.

We understand you. <3

This is really interesting. As someone who's been gradually losing his vision, I've noticed some of this stuff on my own, but it's nice to see a good writeup so one doesn't have to figure out everything from first principles.

Just one more thing to add to my bag of tricks.

I’m wondering if we use it more than we think. When I have my headphones in with noise cancelling on I’m more inclined to bump something in my house I normally wouldn’t or misjudge a distance etc. Pretty crazy to think about forming a subconscious, audio-based world

it's somewhere between silly and mischievous that AtlasObscura provides cute visual drawings yet fails to provide acoustic examples.

I'm curious if people with aphantasia would also be able to create a "mind map" of the area around them

I have tinnitus and can't hear much in a range around 4.5khz... I wonder if this would still work for me

I also have tinnitus. I need to get my hearing checked. I have an appointment coming up.

Teach yourself to echocolate

(2018)

I'm so glad this is surfacing. I remember reading about the skepticism scientists had about this only for them to be finally proven wrong.

My anecdotal experience is that we are so out of touch with our bodies these days that we routinely underestimate just how adaptable we truly are if we have the will or need to learn. So I get frustrated when very useful things like echolocation are suppressed by ignorant and cynical scientists who are unaware of their blind sides because they think they studied hard and read a bunch of papers.

Our realities are shaped by our own experiences but what is sad is when people then shape other people's realities based on their own skewed realities.

I'm glad that the internet is so good at spreading disparate, niche and folky knowledge and forcing scientists to reconsider their priors more often.

Why am I seeing so many comments disparaging "lofty elite scientists" lately? Every scientist I know is a very curious person, more willing than most to listen to the most outlandish ideas people come up with.

Reply on news.ycombinator.com