If someone wanted to reproduce a sound without any electronics in a small space and using a finger nail could it be possible?

For example there are roads that have sounds embedded in them that would play as you drive over them.

If you look at digital sound equipment, you can see that in most hardware the minimum recording value is 8 kHz or 8,000 samples per second.

So if you look at a fresnel lens you could run your finger across it and make a scratching sound. You are hearing it in your head right now.

If you could take an audio sample, lower the sample rate, and 3D print the grooves couldn’t you make different audio sounds that could be played with your finger nail? Ignore the fact that 3D printers that cannot print that fine.

Records have tracks that contain grooves that a record player amplifies. So shouldn’t this be possible?

  • $\begingroup$ You are asking many questions here, some of which are disjoint. Please try to focus on what you actually want to know. $\endgroup$ Dec 6, 2021 at 16:44
  • $\begingroup$ @CarlWitthoft Will try to focus it more. I want to know if it’s possible to create sound that is perceptible and how many samples that would take. It could be two questions but the questions are directly related and don’t make sense without the other. $\endgroup$ Dec 6, 2021 at 19:33

3 Answers 3


Can you have a lower sampling rate? Yes.

Wll it perfectly reproduce the original sound? No.


The problem is with the audible frequencies to the human ear. A good human hearing goes from about 30Hz up to 20 kHz.

If you capture at 8 kHz due to the Nyquist theorem you can only reproduce accurately about 4 kHz. All the other frequencies above 4 kHz will contribute to the reproduced signal in unexpected manner depending on their phase and frequency content.

In a very specialized application you could use the 8 kHz if you wanted. However most audio players don't have the option because for speech it doesn't make sense.

  • $\begingroup$ I usually take Nyquist for the flu but it doesn’t always work. I’d like to reproduce a sound using grooves on a physical object. So that’s why I’m asking about minimum number of samples (grooves/vibrations?) that would need to be created. $\endgroup$ Dec 4, 2021 at 17:43
  • 1
    $\begingroup$ If I understand correctly you want to replicate the effect of a musical road. If that's the case it depends on the sound that you are trying to reproduce. You'd need to find the frequency content and from that determine the number of grooves/vibrations. $\endgroup$
    – NMech
    Dec 5, 2021 at 7:59
  • $\begingroup$ yes, correct. i want to replicate the effect of the musical road. except i have not chosen a specific sound. it could be any sound. i would think the duration would contain the same number of samples regardless of the sound but i guess there could be less samples if there was silence. i’d like to recreate half a second to one second of sound (any sound) on the surface a small physical object and then use my finger nail movement across that surface as the speaker $\endgroup$ Dec 5, 2021 at 8:27

The digital sampling rate must be greater than twice the highest audio frequency.

So if you record with an 8000 rate, you can capture only up to 4kHz frequency.

CDs and Youtube use 44.1kHz standard.

Here is a Youtube video that compares some LP and digital recordings. It can shed light on some questions of the sort you ask. source

  • $\begingroup$ recording double the rate because of stereo? vs mono? or for other reasons? $\endgroup$ Dec 5, 2021 at 8:32
  • $\begingroup$ @1.21gigawatts Please read the answers and comments, and then go look up Nyquist Limit. That will help you avoid asking silly questions ("stereo vs mono" ) $\endgroup$ Dec 6, 2021 at 16:43

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