My friend from physics who doesn't know whistling says he is good at whistling, but the resonant frequencies are ultrasonic so others cannot hear it. Could this be right, or is he just bluffing?

Excuse my unscientific language here, but when you whistle, initially there is just a hollow windy sound, and then you suddenly feel a 'vibration in your mouth' when the resonance 'kicks in'. I think people who don't know how to whistle are stuck in this "hollow" windy sound zone, and the resonance 'hasn't kicked in'. What my friend is describing could be right, maybe the hollow windy sound is whistling; but it's in an inaudible range. What we perceive as the 'resonance kicking in' could just be an effect of when the resonant frequencies enter the audible range.

Short and to-the-point answers are welcome.

I acknowledge the fact that questions regarding human perception are difficult to answer. But I think even if you leave the human component out of this question, it's something worth thinking about. Like, people who play wind instruments will be aware of this very same effect. What exactly is happening when you move from "whistling" to "windy sound"?

  • 1
    $\begingroup$ I'm reminded of an anecdote about Rabi. Upon being introduced to the cavity magnetron, he asked how it worked. "Ah, Isidor, it works just like a whistle," one physicist responded brightly. "And how does a whistle work?" queried Rabi. He was met with silence. $\endgroup$
    – NLambert
    Feb 10 at 0:43
  • 2
    $\begingroup$ Emperor's new clothes. Besides, how does he know the frequencies are ultrasonic? Does he also claim to have ultrasonic hearing? $\endgroup$
    – DKNguyen
    Feb 10 at 1:36
  • $\begingroup$ Pragmatic test: are dogs in vicinity visibly startled when he "whistles ultrasonically"? $\endgroup$ Feb 10 at 20:11

2 Answers 2


You can find everything beautifully explained in this website on brass instruments: https://newt.phys.unsw.edu.au/jw/brassacoustics.html

A short version is that for ultrasonic (>20 kHz) vibrations you would need to increase the air pressure made by the mouth. That can be done by either putting more air through the mouth or making the volume through bigger. Alternatively, you could think of the "embouchure" being very tiny to make the pressure very high (I will do some quick math here and assume a linear relationship between the pressure and with a typical value of 1kPA to make a trumpet sound around 1kHz, maybe 20 times the pressure will lead to ultrasonic soundwaves).

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8313787/ Here you can see that the damage threshold of the tympani is around 30-100 kPa, so I would speculate that you would need to apply such an absurd amount of pressure to whistle ultrasonically that you would rupture your timpani at the same time.

I am neither a clinical expert or acoustics expert, so those numbers are just from "fun" calculations based on the linked evidence, but I hope there are more publications about what is the true range that a person can actually achieve "humanly".

EDIT: It has come to my attention from the comments that I did not clarify precisely what I meant with my analogy to brass instruments.


The wikipedia article is deemed to deep for it being a wikipedia article, but it has some interesting aspects of how whistling occurs in humans:

First, it is defined as a fluid dynamics problem:

The physical theory of the sound-making process is an example of the application of fluid dynamics or hydrodynamics and aerodynamics.

Regarding human whistling, the wikipedia article is confusing but a study from Wilson et. Al. https://pubs.aip.org/asa/jasa/article-abstract/50/1B/366/745861/Experiments-on-the-Fluid-Mechanics-of-Whistling?redirectedFrom=fulltext points to a similar behavior when using a model based on a cylinder with different holes (for input and output).

While it is noted that the fact that I used the trumpet/brass as an example being a farfetched example, as I find further examples where the explanation is done more through fluid dynamics I think it is more difficult for me to make an informed assessment, but I still think that the amount of airflow to have supersonic sound out of a person is unlikely.

  • 5
    $\begingroup$ I would think that the physics of whistling is quite different from the physics of brass instruments. In reed instruments you have a vibrating reed, in brass instruments your lips are the reeds, but in recorders and flutes the air itself (or its flow) is the reed. Anyone who has tried to play these types of instruments knows that you apply a much higher pressure to a trumpet than to a recorder, even for playing the same note. There are many types of whistling, but I would think that most of them are more like playing a recorder than a trumpet. $\endgroup$
    – jkej
    Feb 9 at 19:42
  • $\begingroup$ Whistles are totally unlike brass instruments. If this information applied to whistles, dog whistles could not be used by humans. $\endgroup$
    – Edward
    Feb 10 at 20:16

The claim is almost certainly false.

The normal range of human whistling is from 2-4 KHz. For someone to whistle in the ultrasonic range would require him to reduce the space between tongue and palate by 80% more than people actually do, as that's where the pitch of whistling (and the formant frequencies of vowel sounds) is (are) formed.

Unless he has a mouth the size of a bat, of course....

  • 1
    $\begingroup$ Surely most people can reduce the space between the tongue and palate to 0, right? How is that unbelievable? $\endgroup$
    – Edward
    Feb 10 at 20:08
  • 1
    $\begingroup$ It's possible that some other whistling technique exists. When I had braces, I could do a whistle at roughly 10kHz. I just did a 13kHz whistle using a similar technique- the resonant space is between the lips and front teeth. $\endgroup$
    – Edward
    Feb 10 at 20:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.