Is it possible to play a sound so briefly, during such a short period of time, that it isn't noticed by humans?

I have the idea to use sound for distance measurements. This is common with ultrasound, but I want to do it with common smartphones and thus in the ordinary audible frequences they are capable of generating and detecting. Ultrasound is prefered because it doesn't disturb people, so I wonder if audible frequences could be made inaudible by being brief.

If I make a phone generate a tone during only a millisecond once every five seconds, would it be perceived?

  • While, as Rory says, it's not possible on grounds of just keeping the test noise short, there are actually quite a lot of options you have that might get you there. For starters, you can basically use ultrasonic: the range from 18 kHz to 22 kHz is essentially inaudible for most people, yet is covered pretty well by decent mobile-phone speakers and microphones (albeit with significant phase distortion). Other options might be a good topic for, hm... maybe dsp.stackexchange.com, or perhaps arduino.stackexchange.com. Commented Mar 7, 2014 at 13:21

1 Answer 1


The simple answer is no.

If you make air move, in the audio range of humans, and not too quiet, then there is no duration short enough that it can't be heard.

And a millisecond is a very long period of time for the human ear to react to a sound.

Your options are:

  • use ultrasound, as you mentioned
  • use very low volume (may or may not work for you)
  • accept that you may just have to disturb people
  • Sounds reasonable (no pun intended). We are very sensitive to sounds. Some other of my options are to use enjoyable sounds, target deaf people, wait until phones are so fast that they can use sounds on nanosecond level (because there has to be some limit!?) Or seriously, to look into what inaudible frequences the microphones and speakers of phones can work with. Frequences which only disturb kids who still have perfect hearing :-D
    – LocalFluff
    Commented Mar 7, 2014 at 10:40
  • nanoscale is still not going to make a difference - if the impulse isgoing to be usable by bouncing off a wall, it will be picked up by the human ear, unless it is ultrasonic
    – Rory Alsop
    Commented Mar 7, 2014 at 10:44
  • I should've explained that I would like to measure the distance between phones which send unique signal sounds to each other. Not bouncing sound waves with objects like a sonar, I think that would require sound to be directed in a way which ordinary phones cannot. The phones would in my idea register exactly when they send and receive the signals and the timing data would be coordinated through radio network communication a moment later. I might post more questions here about creating efficient sounds for that purpose (even accepting them to be auible), if I don't find an early showstopper.
    – LocalFluff
    Commented Mar 7, 2014 at 10:58
  • I should add that the reason I would use sound instead of radio is the slow speed of sound waves, relative to computational speeds, which allows for good precision in distance measurements with simple time-of-arrival methods.
    – LocalFluff
    Commented Mar 7, 2014 at 11:00

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.