In the late 80s I worked in a low cluster of buildings, each of which was topped with a band of vertical ridges spaced about 4" apart (sort of like a corrugated roof, but with vertical corrugations). One day a thunderstorm came through, and we discovered that the pulses of thunder, when they hit the corrugations, reflected as a quickly falling tone. The corrugations were working as an acoustic diffraction grating, with different frequencies reflecting in different directions.
_spduchamp 3 hours ago [-]
I love corregated concrete! Those highly ordered early reflections can make a cool sound if you stand about a few feet away and clap your hands. I've wondered if it was possible to vary the bands to encode a designed sound in the reflection. Probably.
davidmurdoch 1 hours ago [-]
Kef has a "metamaterial" they add to their speaker drivers to absorb specific frequencies in order to modify the frequency response curve of the drivers themselves, as opposed to only trying to correct via the cabinet. https://us.kef.com/pages/metamaterial
That is a very wild shape indeed. I wonder if there is some analytical way to derive these or if some type of search algorithm will remain the best way.
ttoinou 10 hours ago [-]
Thats crazy ! Thanks for sharing
Peteragain 8 hours ago [-]
The linked original paper is readable and answered many questions. They simply embrace the idea that some "crazy shape" will work, and then do "machine learning" in a simulation to find it.
amelius 5 hours ago [-]
Cool, but I'd rather see a nicer shape, like a human cochlea, which also splits sounds into frequencies.
Perhaps if they added some more constraints (e.g. on smoothness) they would end up with a shape like that.
WJW 5 hours ago [-]
It seems plausible that Mother Nature had to make some compromises when designing the cochlea, like how much easier it is to grow a (rather symmetric) cochlea than the weird shape from the article and maybe optimizing more for certain sounds than for others.
3d printers and search algorithms don't have that restriction and can directly optimize for optimal splitting and minimal acoustic losses.
amelius 4 hours ago [-]
But is allowing an irregular shape not more likely to get you stranded in a local optimum?
viraptor 2 hours ago [-]
Why? And why wouldn't you expect an irregular shape to be the global optimum?
failrate 3 hours ago [-]
IIRC it is an array of cilia of different sizes that filters sound into different frequencies in the human ear.
xattt 4 hours ago [-]
What sort of ML method would do shape optimization?
voidUpdate 3 hours ago [-]
Quite a lot could probably do it. Have it adjust positions/shapes of blocks placed in a virtual setting, have its fitness function be based off how split apart the frequencies are, press train and make a coffee. Sort of like this https://rednuht.org/genetic_cars_2/
zharknado 12 hours ago [-]
Brainstorming applications of knowing your angle relative to a point source:
- adaptive sports for visually impaired players like beep baseball?
- robot swarm members knowing their relative 2d position with a single microphone? (frequency for angle, amplitude for distance)
- a cheap, durable way for human workers to track the rotation cadence of slowly rotating machinery?
xattt 4 hours ago [-]
Before practical applications, my first thought was a grade-school science fair entry. It’s novel enough for judges not to have seen it for a couple of years.
I wonder how they discovered that "clicking" works, seems so counterintuitive to discover. Though it's fair that I don't spend anywhere near that much time listening carefully and noticing how sharp sounds reflect.
viraptor 2 hours ago [-]
> I wonder how they discovered that "clicking" works
I won't think it's they unexpected - have a walk a forest with loud insects. You'll hear a lot of interesting noises shaped a lot by what you stand near to. Large trees especially change how you hear things quite a lot.
jrklabs_com 1 hours ago [-]
Can it be scaled up and done in reverse to create a point source of coherent audio from multiple speaker components covering different frequency ranges?
muxator 8 hours ago [-]
Neat method. However, the frequency range for the device is 7600-13600 Hz: less than an octave.
sfink 1 hours ago [-]
Ah, so you're pointing out that you could scale it up and down and stack them to convert white noise into chords? Cool!
viraptor 2 hours ago [-]
This is the first device of that kind. It's a research paper. The method described works. They weren't going for the limits of the technology here, but proving that it works. Others with different requirements of frequency ranges can create their own versions.
xattt 4 hours ago [-]
Terrible! I was expecting at least a 7599-13601 Hz split.
egypturnash 15 hours ago [-]
okay who wants to build a musical instrument that works by beaming white noise at a bunch of these things, with some way for the user to rotate them quickly and accurately
catlifeonmars 14 hours ago [-]
I’m wondering if you can change the shape in such a way that rotating one would produce an arpeggio.
wizardforhire 12 hours ago [-]
I’m game to do some heavy lifting if you’re serious.
aa-jv 6 hours ago [-]
I'm in.
Although I think this also has applications as a mechanical control mechanism, replacing expensive potentiometers with relatively cheaper 3D-printed parts. I've dreamed-up an optical/laser variant in my head a hundred times, fun to look at it in the audio domain.
CommenterPerson 4 hours ago [-]
I'd like to passively divert annoying or redundant airport announcements down the drain.
More seriously, great concepts for architects to scale up and control noise?
chrisweekly 12 hours ago [-]
This is the kind of thing that keeps me coming back to HN more often than I should. So cool.
Leo-thorne 5 hours ago [-]
I used to think you had to rely on circuits or resonance chambers to split sound frequencies, so seeing this done with a 3D-printed structure really changed how I think about it.
It made me realize how little design attention we actually give to sound. Most of the time we just passively listen, without thinking about shaping the path of sound. If structures like this can be made smaller and more portable, I can see a lot of interesting use cases in phones or compact voice-controlled systems.
amelius 6 hours ago [-]
Is this similar to how the human ear turns signals into frequencies?
I suppose the thing is linear? So it behaves like a Fourier transform?
adornKey 3 hours ago [-]
I think the ear relies more on resonance (unrolled it's like a stretched rubber triangle). The higher you go, the more it reacts on higher frequencies. This one here seems to just use passive reflection to get the effect.
neuroelectron 16 hours ago [-]
One more step toward building the pyramids.
voidUpdate 7 hours ago [-]
I mean we already built the pyramids...
Cthulhu_ 4 hours ago [-]
Yeah I'm not sure what the grandparent refers to. They're not mysteries or impossible to build, just labor intensive and expensive. Humanity has built bigger structures since then; skyscrapers, dams, underground complexes like the LHC, stadiums, bridges, cities (non-organically grown), etc.
voidUpdate 3 hours ago [-]
There are fringe theories that the pyramids were built using sounds to levitate the stone blocks, which are backed up with footage of ultrasonic acoustic levitation of polystyrene balls, and these days probably ai generated video too. Instead of using the Nile to transport them and ramps and rollers to lift them
bix6 12 hours ago [-]
How the heck do you arrive at such a crazy shape wow this is amazing.
WJW 5 hours ago [-]
1. Start with a non-crazy shape that sort of does the same thing but worse (ie non-linearly or with big power fluctuations for different frequencies). Almost all shapes will interact with a sound wave in a frequency dependent manner so you can quickly scan through many initial options.
2. Define a goal function describing the desired outcome (like the "rainbow" shape of the linked article)
3. Permute the shape you started with iteratively with algorithms like simulated annealing, using the goal function from 2 as a means of defining the quality of the current solution.
The actual scientific paper (https://www.science.org/doi/10.1126/sciadv.ads7497) is worth reading. The goal function is the "Figure Of Merit" (FOM), described in equation 1. They also make a two-prong version they call the "lambda emitter" that takes in a mix of sound frequencies and directs low and high-frequency sound waves in different directions.
poulpy123 8 hours ago [-]
So it's like a prism for sound ?
mhb 3 hours ago [-]
Just like it says in TFA. Go figure.
Onavo 10 hours ago [-]
Solid state FFT?
WJW 5 hours ago [-]
Do you mean a prism?
fellatio 9 hours ago [-]
It would need to do something like spit out each frequency in a different direction then you use light material in circle that flaps when sound energy is transmitted in that direction. Sounds possible. Analog computer of sorts.
ttoinou 10 hours ago [-]
That does seem like witchcraft
bobmcnamara 14 hours ago [-]
Whoa it's like an ear but for light!
recursive 11 hours ago [-]
But for sounds
ttoinou 10 hours ago [-]
Seems like you are filtering the OP white noise broad band joke into a coherent explanation
Perhaps if they added some more constraints (e.g. on smoothness) they would end up with a shape like that.
3d printers and search algorithms don't have that restriction and can directly optimize for optimal splitting and minimal acoustic losses.
- adaptive sports for visually impaired players like beep baseball?
- robot swarm members knowing their relative 2d position with a single microphone? (frequency for angle, amplitude for distance)
- a cheap, durable way for human workers to track the rotation cadence of slowly rotating machinery?
Bloody hell, I can't do all that...
I wonder how they discovered that "clicking" works, seems so counterintuitive to discover. Though it's fair that I don't spend anywhere near that much time listening carefully and noticing how sharp sounds reflect.
I won't think it's they unexpected - have a walk a forest with loud insects. You'll hear a lot of interesting noises shaped a lot by what you stand near to. Large trees especially change how you hear things quite a lot.
Although I think this also has applications as a mechanical control mechanism, replacing expensive potentiometers with relatively cheaper 3D-printed parts. I've dreamed-up an optical/laser variant in my head a hundred times, fun to look at it in the audio domain.
More seriously, great concepts for architects to scale up and control noise?
It made me realize how little design attention we actually give to sound. Most of the time we just passively listen, without thinking about shaping the path of sound. If structures like this can be made smaller and more portable, I can see a lot of interesting use cases in phones or compact voice-controlled systems.
I suppose the thing is linear? So it behaves like a Fourier transform?
2. Define a goal function describing the desired outcome (like the "rainbow" shape of the linked article)
3. Permute the shape you started with iteratively with algorithms like simulated annealing, using the goal function from 2 as a means of defining the quality of the current solution.
The actual scientific paper (https://www.science.org/doi/10.1126/sciadv.ads7497) is worth reading. The goal function is the "Figure Of Merit" (FOM), described in equation 1. They also make a two-prong version they call the "lambda emitter" that takes in a mix of sound frequencies and directs low and high-frequency sound waves in different directions.