I figured out a way for people to test the audibility of ultrasonics (controlling for IMD)

[vintologi]

In some studies testing the impact of ultrasonics they used a separate "super tweeter" with digital crossover to avoid IMD from the ultrasonics interacting with lower frequencies.

But there is actually a stupidly simple way for people to test that without any special equipment (besides speakers and electronics able to reproduce the ultrasonics). You simply convert the track you want to test to mony, downsample down and up (or just use a lowpass filter) with audacity and then play the missing ultrasonics from the other speaker placed as close as possible to the other speaker.

Note that there can in theory be audible distrortion from interactions inside some bad amplifier (due to channel interactions) but that will not be anywhere close to audible if you have a solid amplifier. You can of course use two monoblocks if you are paranoid but that seems like a massive overkill to me.

I did make a compilation with some tracks earlier here:

High resolution audio

CD quality allows you to reproduce frequencies below 22.05 Khz at a a signal to noise ratio of up to 96 dB, this is actually pretty good considering CD was format was released in 1982. It's actually very difficult to fully utilize the Signal to Noise ratio CD offers even today The topping...

vintologi.com

I did get the impression that there was an audible difference for a short section for one of the tracks. Said section also didn't have particularly strong ultrasonics making audible IMD very unlikely with my current tweeters (TW29BN). Still just to be sure i made a version where the by themselves inaudible ultrasonics are placed in a separate channel (frozen-short-A vs frozen-short-B)

33.47 MB folder on MEGA

7 files

mega.nz

I don't think i will bother moving one of my speakers when testing this.

Let me know if there is some good test i can do on my current amp to rule out audible crosstalk distortion from the ultrasonics (i should have upgraded to something much better years ago).

 

[respice finem]

What would be even more simple: apply a high pass filter to eliminate all frequencies under 20 kHz and listen if you hear anything via the "supertweeter".
Amateur prophecy: There will be silence...

 

[whazzup]

I don't understand.... I play a youtube frequency sweep video on my speakers and there's a very, very clear cut-off in audibility the moment it hits ~15,000hz. Like someone turned off my speakers. That's not a sufficient test? Or is it a research attempt to come up with an experiment that proves or disproves people can somehow sense ultrasonics?

 

[vintologi]

I don't understand.... I play a youtube frequency sweep video on my speakers and there's a very, very clear cut-off in audibility the moment it hits ~15,000hz. Like someone turned off my speakers. That's not a sufficient test? Or is it a research attempt to come up with an experiment that proves or disproves people can somehow sense ultrasonics?

The point is to see if ultrasonics can still make an audible difference when mixed with the lower frequencies from the other speaker.

It is worth noting that playing ultrasonics from a separate tweeter will create some interference issues depending on how far the speakers are apart.

 

[vintologi]

What would be even more simple: apply a high pass filter to eliminate all frequencies under 20 kHz and listen if you hear anything via the "supertweeter".
Amateur prophecy: There will be silence...

That's intended since the purpose is testing ultrasonics that are inaudible by themselves.

I plan on testing this blind 10 times

9 or 10 correct: test ends right away.
7 or 8 correct: test get extended.
More than 3 wrong: test will be aborted.

 

[vintologi]

Update: i got the impression that moving the ultrasonics to a separate speakers made it harder to hear a difference.

So no guess and speakers moved closer to each other to see if that makes it easier. Test will continue.

update2: not sure if changing the position of the speakers helped (they are still rather far apart).

If you can only hear a difference when the ultrasonics is played in the same channel it might simply be due to IMD but i did not think that would be audible with very high end tweeters like the TW29BN. I did pass a 24-96 vs 16-44.1 blindtest many years ago but 16-44..1 actually sounded better which is a red flag (my hearing back then was also significantly better which probably helped).

I have been concerned about audible IMD from the amplifier since the one i bought isn't particularly good (it's a cheap class D one limited to like 10W).

I have not however passed a proper blind-test with regard to sample rate with my current tweeters (i did try 24-96 vs 24-54 first which was aborted due to having been too hard in addition to concerns about the test setup).

 

[Redacted]

I was playing a track from "Hunky Dory" (an old record I know, from before the days of electricity) with a view to doing a version of on my guitar - but blow me, those pesky ultrasonics made my fingers freeze up.

 

[vintologi]

In terms of difficulty i would say that frozen-short-A vs frozen-short-B is at least 20 times harder to differentiate in a blind test than flac vs ogg vorbis Q100.

I might need do 1 of the following (at least)

0. Get a better amplifier with wider ultrasonic coverage.
1. Reduce the sampling rate tested to 44.1 KS/s.
2. Improve my hearing (not sure if that's even possible to begin with).

I could also try testing a different sample and see if that's easier (the ultrasonics were more intensive elsewhere for that song and there are other songs to consider as well some of which have better looking spectrals).

 

[vintologi]

One issue with using a separate tweeter for the ultrasonics is that it will cause interference issues at the crossover area and you end up having to have your head in an exact position for it to work well.

Another approach is to have one speaker always play the the track with the ultrasonics intact and have the other speaker play just the ultrasonics at inverted phase.

Here is one example you could try on your own (harp music with significant ultrasonics):

https://mega.nz/folder/5LNFCRKS#S7IosmMPoliJVslPFOOKrw

Full Album:

https://gkmharp.bandcamp.com/album/suitzer-harp

Let me know if you want me to make a test-track for some specific song for you.

Short clip with guitar:

MediaFire

MediaFire is a simple to use free service that lets you put all your photos, documents, music, and video in a single place so you can access them anywhere and share them everywhere.

www.mediafire.com

 

[Spkrdctr]

As expected you will get one of my old crazy experiments from over 35 years ago. I was wondering at the time if the speakers I had could play above 20khz. I got a test tone of 20khs to 25khz which when played at very loud volume was dead quiet to me. No sound at all. I could hear nothing. BUT my cat would run from a different room come ripping into where the speakers were and put his paws up on the speaker and put his nose dead on the tweeter! I repeated this about 20 times over a few months. I don't know what he thought the sound was but he always exploded with speed to put his nose directly (about and inch) off of the tweeter center. So, that is my old man story. Thought you might enjoy hearing about our cats and dogs and how they really can hear so much that we never hear. We are sound handicapped compared to them........

 

[fpitas]

As expected you will get one of my old crazy experiments from over 35 years ago. I was wondering at the time if the speakers I had could play above 20khz. I got a test tone of 20khs to 25khz which when played at very loud volume was dead quiet to me. No sound at all. I could hear nothing. BUT my cat would run from a different room come ripping into where the speakers were and put his paws up on the speaker and put his nose dead on the tweeter! I repeated this about 20 times over a few months. I don't know what he thought the sound was but he always exploded with speed to put his nose directly (about and inch) off of the tweeter center. So, that is my old man story. Thought you might enjoy hearing about our cats and dogs and how they really can hear so much that we never hear. We are sound handicapped compared to them........

I prefer to think we have the proper band-limiting lol. He probably thought that was critter scratchings, and was ready for some murder.

 

[LiveTek]

As expected you will get one of my old crazy experiments from over 35 years ago. I was wondering at the time if the speakers I had could play above 20khz. I got a test tone of 20khs to 25khz which when played at very loud volume was dead quiet to me. No sound at all. I could hear nothing. BUT my cat would run from a different room come ripping into where the speakers were and put his paws up on the speaker and put his nose dead on the tweeter! I repeated this about 20 times over a few months. I don't know what he thought the sound was but he always exploded with speed to put his nose directly (about and inch) off of the tweeter center. So, that is my old man story. Thought you might enjoy hearing about our cats and dogs and how they really can hear so much that we never hear. We are sound handicapped compared to them........

I had to drop my processing down to 48khz because apparently my amps have no problem going way past 20khz. My cats have been acting weird around the tweeters lately.

Did a measurement in rew and sure enough I had extension past 20khz. Tweeters are sb26adc.

 

[vintologi]

I had to drop my processing down to 48khz because apparently my amps have no problem going way past 20khz. My cats have been acting weird around the tweeters lately.

Did a measurement in rew and sure enough I had extension past 20khz. Tweeters are sb26adc.

View attachment 552655

Note that your mic probably starts to roll off past 20Khz.

 

[LiveTek]

Note that your mic probably starts to roll off past 20Khz.

Mine goes all the way up to 50khz.

 

[vintologi]

Mine goes all the way up to 50khz.

Then it might the the amp limiting the ultrasonic response.

 

[MAB]

Then it might the the amp limiting the ultrasonic response.

Actually your ears and your human perception limit the ultrasonic response.

Nobody was able to reproduce any of the Oohashi hypersonic results.

I don't think a casual attempt will get results. Likely the effect doesn't even exists and it was experimental error, or worse. But knock yourself out. I have owned super tweeters, even really nice ones, even when I was young and might have had a chance of hearing them. The one enduring thing that fooling around with them taught me, the content above 15kHz is not nearly as important as audiophiles believe.

They are also very expensive, consistent with many other audio devices that have no value in reproducing music.

 

[kyuu]

One thing I never see addressed in these discussions of ultrasonics is preference. Even if one can hear these frequencies over 20kHz, and there's actual content of note that's not simply noise (both dubious propositions on their own), is hearing those frequencies even desirable? Just based on how ear-piercing a 15kHz tone is (the highest I can reliably detect at this point in my life), I'm not sure it would be a positive contribution.

 

[dualazmak]

We can find so many similar discussions on the same theme within ASR Forum....
Here, let me share "again" my practical "summary of stance and policy" which I shared in my post #532 on my project thread.

 

[fpitas]

One thing I never see addressed in these discussions of ultrasonics is preference. Even if one can hear these frequencies over 20kHz, and there's actual content of note that's not simply noise (both dubious propositions on their own), is hearing those frequencies even desirable? Just based on how ear-piercing a 15kHz tone is (the highest I can reliably detect at this point in my life), I'm not sure it would be a positive contribution.

I tend to agree. Plus, I've seen tests where ultrasonic tone intermod was audible as distortion products in the audible band.

 

[vintologi]

Nobody was able to reproduce any of the Oohashi hypersonic results.

That's misinformation. There have only been a few replication attempts and results are mixed.

We have this short paper which may have found something but nothing particularly convincing:

http://www.jslst.org/documents/Conference/2015/data/thesis/1e2-04.pdf

So far i have only found 2 decent quality independent replication attempts (with regard to brain activity) and seems like neither bothered to minimize the impact of IMD by playing the ultrasonics from a different speaker:

https://pmc.ncbi.nlm.nih.gov/articles/PMC5285336/

https://www.nature.com/articles/s41598-024-57749-w