（评论）

（评论）
(comments)

原始链接: https://news.ycombinator.com/item?id=40624709

在东京，有神秘的防徘徊装置会发出高音调的铃声。这些设备由于其微妙性而常常被忽视，通常位于室外建筑入口或公共区域附近。尽管美国也有类似的设备，但它们经常发出脉冲而不是连续发出声音。日本的变种会产生一致的高频嗡嗡声，范围约为 18-20 kHz。许多人，包括提交人，都表示在遇到它们时感到不安。一些理论表明，这些设备针对的是年轻人，利用他们比老年听众感知更高频率范围的能力。尽管尝试识别这些设备，但文档有限，这使得精确表征具有挑战性。然而，根据现有证据，这些有趣的声学现象有助于阻止徘徊，同时为城市环境带来神秘元素。

FM Radio transmitters broadcast a continuous 19 KHz pilot tone to indicate that it's a stereo broadcast. If it's present, then there's also a 38 KHz AM modulated carrier, which has the stereo information.

The main FM modulation is (Left + Right) audio channels - the baseband. The 38 KHz subcarrier is (Left - Right). Notice that the stereo information is likely to be higher noise, being AM and without a lot of power. So on weak FM broadcast signals, you're often better off setting your tuner to mono.

Some cruddy FM tuners don't properly filter out the 19Khz pilot tone, resulting in ear-ringing in the listener.

The unfiltered pilot tone was a problem when recording FM radio onto tape, beating against the bias tone and confusing Dolby NR into thinking there was constant sound.

High-end cassette decks had an "MPX Filter" which was a sharp 19kHz filter before the recording system.

You betcha!

And FM broadcasts have a high-frequency pre-emphasis, which adds something like 10dB of boost to audio high frequencies. This is to reduce high frequency noise that occurs in analog FM systems. Radio receivers have a matching de-emphasis high-frequency cut.

Indeed, both vinyl phonographs and audio tape systems employ similar pre-emphasis and de-emphasis. It's the RIAA curve for phonographs. Not sure what the standard is for mag tape.

Very interesting. I can remember being able to hear a high pitched tone when I was 4-6 shopping with my mother in department stores that seemed to get louder when we approached flush mounted speakers in the ceiling. This article measures the frequencies and explains it's a normally operating feature of the systems.

Going into Sears used to drive me nuts. My parents finally accepted that I was actually hearing something that they couldn’t, but it took a while.

I don’t know if it was the loss prevention system that was doing it, or what. All I could do was localize it down to being anywhere remotely close to the entrances. And the closer you got, the louder it got.

These days, my tinnitus is constantly ringing at about that same frequency. So, I would probably never be able to hear it again.

Exactly this. When I was 17 I worked at a JC Penny store as a stock boy. Every time I approached certain entrances (like the main one that opened into the shopping mall) and other areas of the store my ears were assaulted with a high pitch squeal. Never figured out what it was, nobody else seemed to hear it. I've always had tinnitus, and the frequency always seemed to be the same. In fact, I'm listening to it now.

Have seen a few stores and public places installing speakers pushing out a high pitch tone to irritate youth so they won't hang out where they aren't wanted. Was it some trial run on this tech maybe? :-)

It's designed to be just outside of hearing for adults so only kids and youth can hear it. Did a simple test with our daughter and she can hear quite a bit higher frequencies than we can so apparently it's a thing. Except I can hear higher frequencies than my wife so not very useful I guess.

Could have been the pulse frequency of old magnetostriction loss prevention rfid tags.

The tone is not meant to be audio, but might be inadvertently emitted by the electronics.

I wear hearing aids and could always hear them as I passed near them. Obviously not the same as real sound as it was just interfering or something.

Or the idea that the humans are just blissfully unaware while their 'hyper' cat/dog is stuck inside a home with some device screaming 24/7, LED lights flashing, etc.

To this day, not one person in the spec reviews in whatever overseas OEM electronics or those split AC units (which are fantastic in one regard) reflects on persistent light in a room where people are sleeping? The worst is encountering AC units when arriving at a hotel that do not have a "display" button to turn off the 20C display in white LED. Are the actual people that make these and make those decisions just tougher or they're just ignorant?

Lately we had both our tenants in our other house (on the same lot) leave, my wife trapped two stray cats inside the house, one escaped, and we are trying to tame the one who stayed.

I moved the TV, XBOX, a $50 PC and almost all of my optical discs (but no books) over there because (1) I won't be so mad if a copy of Disney's Frozen gets pissed on (he's a tom), and (2) if it does the disc will be undamaged and at worst I can discard the package.

That room gets really dark and it doesn't have a switch controlling an overhead light so turning off a floor lamp and then sneaking out the door without creating an opening for the cat to escape (though he's much more inclined to hide than run) leaves me quite happy for the little LEDs and stuff that create just enough light to get from the couch to the door.

I think the dogs can hear our induction stove with ultrasonic whine, but somehow my cat will sleep blissfully in a room with a VGA CRT whining from 30-80 kHz (perhaps quieter than the stove), and used to sit in front of a CRT TV whining at 15.7 kHz so loudly I couldn't stand it without headphones.

When I was a kid I noticed that the adults in my life couldn't hear that CRT sound. My son bought a CRT TV with a built in VCR to use with a Nintendo 64 to get out the last frame of latency. I noticed the other day that I can still hear that sound, but I've always had sensitive hearing and I've always kept amplified sound at a low level and worn ear protection while using machinery, going to motorsports events, etc.

(In the past I have volunteered as a sound man for live music events but I have never been at a venue which I thought wasn't too loud in terms of absolute sound pressure or in terms of there being audible distortion in the amplifiers & speakers, so I haven't made a habit of it. My son won tickets to a 38 Special + Foghat show that we went to last week, I didn't mind the loudness much but it was clear to me they were pushing the sound system just a little too hard.)

I remember in school, we'd often hear the TV whine and know we were watching a tape in class that day without needing to look up at the TV (mounted high in the corner by the ceiling.) The teachers used to wonder how we knew and didn't believe us when we said we could hear powered on TVs.

GP's comment said 100Hz because that's what you get from full-wave rectification of 50Hz mains, so they were talking about that. It'd be 120Hz in countries that use 60Hz mains (e.g. in North America).

I'm not sure I ever managed to hear a 19kHz tone, but I'd say it might have been possible (or the tone was a lower frequency one).

But yes the old CRT whine I could definitely hear

I was young. My mother could not hear it and thought I was just being annoying. This is several decades ago, now I can barely hear my turn signal blinker and am preparing to emulate my father's practice of driving a few miles with his blinker on after signalling a lane change.

One thing that can help is pressing the turn signal lever with less force than is needed to latch it. This should cause the indicator to flash three times and then stop. It’s an operating mode I wasn’t aware of for decades.

It's been too long since I posted to edit this, so I'll reply to my own comment: something else my car does that isn't at all unique and some people might be delighted to learn (if theirs supports it too), is that if you hold down the unlock button on the keyfob (rather than press it), it lowers all of the electric windows, and if you hold down the lock button, it raises them. Useful if you left one of your windows open and it starts raining; you don't need to go to the car to shut the windows.

I've been trying to break that particular habit, on the grounds that most of the purpose behind a turn signal is to indicate entry into a new lane rather than departure from the old one. Not sure what the law says about that, though.

Both equally important.

If the signal is used sufficiently in advance (aka "indicating" your lane departure),

and you're not cutting things so close you need a blinker to fend off cars behind,

then there should be no problem turning it off once half of your vehicle is in the lane.

I've been tested to 22+kHz, autistic, and older. It was the worst era.

Some of the awful roars of noise I've heard while older people called me a weirdo. CRTs, early checkout scanners, some audio systems.

And back when "as seen on TV" electronic dog whistle trainers were all the rage? Torture.

A few years ago I went to the beach. The place I was staying at had a CRT television. I tried to use it once. I have no idea if something was wrong with it, but it was loud enough to cause pain. Once I turned it off, everything was quiet again and the pain was gone.

You'll usually stop hearing the CRT whine in your 20s, 30s if you're lucky. It's at nearly 16KHz and the highest frequency you can hear always drops as you age.

Of course, these days people also regularly don't hear it since CRTs are nearly all gone.

Even back in my youth and early adulthood, there was a wide variation on how loud CRTs were between models and individual units. There were some that were close to silent for me, while others were very noticeable.

Most of us don't.

The signal is at a frequency of one cycle per line-second, but there's an idle period while the beam goes back up to the top where it still cycles so the actual frequency will be a bit higher than a simple lines/frame * frames/sec calculation would indicate. Even for broadcast TV that was beyond what a lot of adults can hear (but a lot of kids can, as plenty of posters in this thread have said.) I want to say 15.7khz but I certainly won't swear to that. As your resolution goes up the frequency will go up so some people would be able to hear what resolution a monitor is set to.

I had something similar, when i was young and walking the dog in the evening, i could here a high pitch tone if they had the tv on when i walked by houses.

Has the hypothesis that this is noise from a class-D amplifier (which I assume most commercial speakers use) been eliminated? Those generally switch at MHz frequencies, but have subharmonics all over the place except the ones that have been carefully eliminated by filtering. I wouldn't be too surprised if there was a cheap brand of class-D amplifier that had a 20 kHz tone.

I'd also be suspicious of this. You can also get a lot of beating artifacts from having many amplifiers in close proximity. If the speakers were still larger "mid-range" horns as were seen in decades past, this also might be less of an issue (low efficiency above 10kHz), but for cost/aesthetic reasons the speakers have shrunk quite a bit so they no longer mechanically filter much.

If the problem was something everyone could hear it would likely be fixed, but since it's a minority, it's likely to persist.

One other possibility for active US is crowd tracking. I've seen solutions that monitor # people and walking speed at doorways using ultrasonics. A lot of "one way" doors also use simple US detectors.

I live in Japan, Tokyo, and sometimes I come across these annoying/distracting high-pitched noises even in residential areas. They are a nuisance for bypassers too, if you've got ears young enough. But the residential areas don't have any problem with loitering youngsters, so I keep wondering if the noise is targeted against pests or wild animals. This article made me think if I should seriously start measuring the volumes and frequencies objectively. What kind of equipment would you need to get some data to get started with?

I've encountered this a few times in Tokyo myself; from memory at the very least outside of a mall in Shinjuku and outside of a building a block or two away from the Shibuya Scramble crossing. My traveling companion never noticed them until I pointed the sound out, so for those who find themselves in Japan pay attention if you notice a pressure in your ears and a high-frequency sound that seems to get more/less intense as you walk. If you follow the sound to its source you'll probably find that it seems to be coming from a metal trapezoidal horn, usually mounted from ceilings in places outside buildings from standard electrical junction box with exposed conduit. The sound itself is a constant, high frequency (18-20khz?) ring, and can get pretty intense as you approach the device.

I believe they're anti-loitering devices targeted at young people, who can hear a higher range of frequencies than older people. I've encountered them all over the US (if you're in the Bay, there's one mounted outside the Twin Peaks Tavern in Castro), though usually they pulse rather than emit a constant buzz. They're sporadic but common in Tokyo for sure, I encountered at least a dozen in the combined week or so I spent in various regions there. It was always creepy to me whenever I'd discover one, the sound was slightly nauseating if I ever got too close and left a strong negative impression, as though they were some sort of threat.

If anyone knows more specifics about these things btw, happy to have details. I spent a period of time trying to find the exact device but never could find any pictures of them, and unfortunately neglected to when I was actually in Japan.

I came across some of these in Tokyo recently as well. Super annoying, and I'm 40. I thought I'd be immune by now. I assumed they were intended to drive away cats or rats or mosquitos.

Can't speak for Tokyo, but there are streets in my American suburban neighborhood I avoid because of (presumably) animal-deterrent noises. I'm also 40, and surprisingly can still hear them - and an audiologist had my ears tested last month, and said my hearing is excellent for my age.

Which is a bummer when I have to cycle around several blocks to avoid hearing the squeaks that are physically painful to my ears.

> squeaks that are physically painful

Those may be covered under loud noise regulations, if the official measuring machinery has the cutoff frequency set high enough (no reason why not).

In my experience those regulations are only ever used to shut down parties with loud music. If it's any sort of industrial or non-obvious noise, good luck getting people to care about it. I have bad memories of complaining in an online community about the extremely loud safety beeps from construction vehicles operating next to my home and being told I was a monster because I must want children/workers to die - and that was a sound whose mere existence didn't need to be explained.

Yeah, maybe. But the hassle of trying to convince the town this house is in - which isn't my town - to do something about it is significantly more difficult than just riding my bike around the place.

I've always wondered about that. I'm also bothered when walking around Tokyo (it's very prevalent in Ginza for some reason). I've also seen it in other cities in Japan but I've never had that issue in other cities in Asia (Hong Kong, Shanghai, Bangkok, etc..)

I'm 40 too but as a kid, I was very bothered by the high pitch wine of the tv turning on, so I'm guessing I've always been rather sensitive.

> what kind of equipment

Any microphone hooked up to a laptop with a microphone input. Or a phone running some audio spectrogram app.

Getting started is very easy, but then you'll be constantly tempted by better sensors, more sensors, more channels, more analysis...

I assumed that most of the normal consumer-level stuff is (perhaps very deliberately) insensitive to sounds that human ears are insensitive to? The common codecs, for sure, tend to have rather harsh cutoffs around 20kHz (the common sampling rates being 44 or 48kHz, and the Nyquist frequency half of that, 22-24kHz, so you have to start filtering well before?)

Plus, I'd rather not walk around with a laptop. But perhaps smartphone apps are capable of consuming the mic input unfiltered? (Never tried recording APIs for that purpose)

Normal computer recording only goes up to 48kHz (24kHz Nyquist rate, with analog filters lower than that to prevent aliasing) but high-quality equipment might support 96kHz or even 192kHz.

even a headphone plugged in wrong is a good enough sensor to start.

when it gets bad you start looking at designing your own ribbon elements and considering active pickups. when it gets worse you may find yourself deeply concerned about ADC and clock skew.

And many modern sound card IC sample at an effective rate of 192kHz and use internal DSP to downsample that to the requested sample rate. Much easier to design a good low pass filter on the cheap when your Nyquist frequency is on 96kHz instead of 22kHz.

So the frontend of the sound card might actually be remarkably wide.

it might; depending on your goals: microphones color the sound and their limits can be as valuable as their capabilities, in the right context.

see the mics that harmonica players lust after. "You traded the Cadillac for a microphone? ... I can see that."

I believe many sound chips still have ways around any pee-processing they may apply, might check the folks using them for low rate oscilloscope inputs

There are lots of "rat deterrant" devices which emit varying tones 15-20 kHz. they often emit chirps or frequency sweeps. Sometimes always-on, sometimes with an IR sensor (in battery powered versions).

I work for a manufacturer of PA systems, our amplifiers only do a periodic test every few minutes, one channel at a time. We measure the current the speaker circuit draws, and if that changes drastically we know something is wrong.

Have you ever listened to a badly mixed song with a persistent high-pitched whine undercutting part (or all) of it? Certainly that's not on purpose either, but I doubt anystudio equipment is giving off an intentional "test" whine either, so I've always wondered how that ends up happening.

I've got an example of this: Some Blackpink songs (cannot remember which ones, possibly Playing With Fire or DDU-DDU-DDU) were mastered and released on Spotify/iTunes [unsure if has been removed now] with a 18-19kHz carrier tone (iirc it was actually a number of tones) modulated. The signal was very loud. I first noticed this in a friend's car and assumed it was because the song was ripped from YouTube, but later realized that all of the versions have this.

The reason for this is because at concerts, you can buy a piece of merch called a lightstick. Essentially, it is a STM32 and a mems microphone that is demodulating the carrier signal, so it can flash in time with the music. The tone has to be

I assume the theory goes that anyone who has been to enough concerts to have a lightstick is no longer capable of hearing the noise.

Sometimes it's old ears, sometimes it's the people in the room being oblivious to it from hearing it too much. I tell the story a lot about kind of the inverse of this - I was mixing my band's (terrible) recording, and during a weekend of mixing, I heard the squeak from my cheap kick pedal, like an icepick to the brain, piercing through the mix. I did my darnedest to EQ it out, and felt pretty good about it in the end.

On my Monday morning drive to work, I put on a CD by the band Cake, an album I'd listened to dozens, if not hundreds of times at this point, and would you believe that now, piercing through the mix, I heard the squeak of the drummer's kick pedal. Never noticed it before, but because I was so focused on those frequencies for two days straight, now it was all I could hear.

I switched to listening to purely electronic music that month, and I never noticed squeaky kick drums in studio recordings again.

It's surprisingly common in older music. I assume it's because recording studios used to run SD CRTs and everybody working there had already lost their high-frequency hearing. A 15.68kHz (mid way between PAL and NTSC horizontal refresh frequencies to remove both of them) notch filter fixes the problem in most cases.

Whenever my daughter puts on recent pop music, I can tell instantly. Every pop song singer seems to have their voices artificially boosted into one or two higher harmonics. It's hard to describe in words but instead of the singer singing a single note, it's as if she (and it's very evident with female singers) has some very, very high pitched harmony automatically added, which lasts the whole song.

It's not like a single frequency buzz or carrier wave. It changes along with the pitch of the lyrics and clearly some kind of filtering they're doing to the voices. Either that or I'm paranoid and hearing things.

The other comment is right that autotune is almost always present in recent pop vocals - even when the singer is excellent, a little bit is now commonly added to give it that certain sound.

Having said that, I think what you might be hearing is actually another modern trend - which is just boosting the high frequencies in the vocal a lot. Pop vocals right now are mixed very 'bright', sometimes almost painfully so.

Also common to actually have multiple vocal tracks. Sometimes just a double singing the same thing, but sometimes it can be pitched up or down an octave, and I've also seen "whisper" tracks where a whispered vocal is layered with the main one to give it even more of that bright breathy sound.

> autotune is almost always present in recent pop vocals

Autotune is often present in almost any decently-high budget studio recording of loads of different genres. Its just not obvious.

Autotune doesn't need to sound like Cher in Believe or T-Pain in any song he sings or Kesha. They're just cranking some of the settings to make those super obvious artifacts. When a sound engineer uses autotune (or the many other similar effects) you wouldn't even notice its there.

Speaking as someone who moonlights doing live music production, autotune is used loads of places.

When Cher's Believe came out, autotune had been a thing for several years. Its just that most people thought people didn't like speed == 0 until Cher made it a thing.

I had almost forgotten about the “Mosquito” devices for deterring youths that became ubiquitous in some areas during the late 2000’s and early 2010’s.

They were quite expensive and I know of a few people who got annoyed and specifically vandalised them due to the annoying sound.

I'm glad those seem to have mostly faded away? The installation of them especially in spaces open to the general public has always struck me as deeply sick and antisocial

There’s still a few around where I live, way less than before though and I think it’s old installations that somehow still function instead of “net new”.

What’s awfully irritating is that I can still hear them. I’m in my 30’s with pretty gnarly hearing damage, and can still hear that frequency range.

I searched the article for "occupancy" and nothing. These ultrasonic tones are most likely ultrasonic occupancy sensors used for lighting and/or security. Typically they're 25kHz or 40 kHz (i.e. out of normal human hearing range), but it's possible that they are measuring some subharmonics of the sensors. I've certainly encountered ones that have mechanical issues and will spit out frequencies that I can hear too well.

induction stovetops create extremely loud high pitched noises, some get up to 100+ decibels. You can't "hear" it but at higher heat levels you can definitely "feel" it in your ears.

LED bulbs must drive them crazy. Many emit ultrasounds that I can hear when I'm next to such bulbs, like that of a vibrating capacitor (low quality AC filtering, low quality capacitor, or both).

Totally speculative territory, but I’ve had headaches when going to a shopping center since I was a kid.

It’s infallible, one to two hours and my head is exploding. I’ve always attributed it to the noise, crowds and sensory overload but it never happens elsewhere. Could it be?

What does the author use to measure sound and create those graphs? I'm interested in automatically monitoring an annoying noise that's coming from a neighbour.

As others in the comments have already mentioned, Studio Six makes very nice Audio Tools (at least on iOS/Mac, not sure if they are also on Android). Their SMAART module is the cheapest way to get access to SMAART tools - which can be a thing in and of themselves. I use them to asses what's going on with live sound I occasionally mix. What's really nice is they calibrate fairly accurately to Apple hardware out of the box. More than accurate enough for my purposes.

> .. there's a clear Doppler shift in the reverb. The frequency shift goes from positive to negative at the same moment that the scooter passes us, seen as the wideband wheel noise changing color.. The speed of sound at 15 °C is 340 m/s. The maximum Doppler shift here seems to be 350 Hz. Plugging all these into the equation we get 11 km/h, which sounds like a realistic speed for a scooter.. Automated speed trap in the car park..

Doppler radar, https://en.wikipedia.org/wiki/Doppler_radar, also applies to WiFi reflections, which can be visualized, https://www.youtube.com/watch?v=g3LT_b6K0Mc.

Upcoming 2024 "AI" PCs, tablets and phones with NPUs and WiFi 7 can use machine learning and doppler reflections to infer human activity in 3-D space (proximity, movement, heart rate, gestures), via IEEE 802.11bf WiFi Sensing, https://www.technologyreview.com/2024/02/27/1088154/wifi-sen...

> Soon, thanks to better algorithms and more standardized chip designs, [WiFi] could be invisibly monitoring our day-to-day movements for all sorts of surprising—and sometimes alarming—purposes. Yes, it could track your breathing. It could monitor for falls. It may make buildings smarter, and increase energy efficiency by tracking where people are. The flip side of this, however, is that it could also be used for any number of more nefarious purposes. Someone outside your home could potentially tell when it’s vacant, or see what you are doing inside.

From the description of the product, I think the intention is that the tone is received electrically at the end of a cable run, so receipt of the tone indicates continuity of the speaker cabling. This high frequency has been chosen to ensure that it is low enough to be amplified and transmitted along with the audio content, and is high enough to be attenuated by the tweeters (so little energy is transmitted), and high enough in frequency so that whatever is emitted by the tweeters will be inaudible.

If this is the case, it would sound like there is no need to encode any information on the signal, just it's presence is enough to test the wiring is intact. I'm assuming this is so that if the PA is required for, say, an emergency broadcast, it is fit for purpose.

The presence of a <20Khz signal would imply that it was found that in whatever installation was being used that the bandwidth of the amplifiers or some other filter was attenuating the test tone too much to be received properly, so this bodge was put in place.

When I used to work with PA systems we had this really cool old-school looking impedance meter. We used to check the continuity of the PA like you do with a resistance meter in an electrical circuit.

I dont know much about analog audio but if I am remembering correctly, the device could inject a 1Khz tone that was only heard on speakers with impedance problems.

https://toa.co.id/product/zm-104a-impedance-meter

I'd like a way to use an array of mics and do some sort of visualization and video overlay to try to see and locate the sources of these kinds of high pitch sounds.

> I might have been scrolling through the audio spectrum while waiting for the underground train;

What gadget is this person using for this? And audio-y flipper one of sorts?

（评论） (comments)

（评论）
(comments)