0
$\begingroup$

Why do some times notes take these weird structures (circled below). That is a guitar monophonic audio waveform by the way (3 note melody).

waveform

There are 3 notes played in this audio file. These are the beginning of the waveform (0), around sample 52000 and around sample 95000.

My two questions are:

  1. Why did this weird looking negative amplitude peak get created (circled)?
  2. Right after the final note (where the arrow is), why is there such a big change in amplitude (down, up, down again)?

The audio file is here.

$\endgroup$
  • 2
    $\begingroup$ can you link us to a the sound file somewhere? i wouldn't know what to call that without hearing it. $\endgroup$ – robert bristow-johnson May 17 '17 at 4:56
  • $\begingroup$ Did you listen to it? It's most likely some extraneous noise like a string hitting a fret or something creaking $\endgroup$ – endolith May 17 '17 at 18:27
  • 1
    $\begingroup$ i can't even hear that circled spike. i just hear three plucked notes. $\endgroup$ – robert bristow-johnson May 17 '17 at 18:45
  • 2
    $\begingroup$ Yeah it's ~8 Hz, very low frequency, maybe just from moving the guitar around? A high-pass filter will remove most of it. There's more low-frequency garbage at 25.5 seconds. $\endgroup$ – endolith May 17 '17 at 18:55
  • 1
    $\begingroup$ well, then the OP has an explanation of the phenomenon. $\endgroup$ – robert bristow-johnson May 17 '17 at 19:05
3
$\begingroup$

Here are some possibilities:

Transverse waves on a guitar string have two polarized components, one parallel and one perpendicular to the guitar's soundboard. User hotpaw2 mentioned energy exchange. It is imaginable that energy transfer could happen between the components. The component parallel to the soundboard is largely inaudible, so it could stealthily store energy until released later.

Or it could be that you have two strings with modes that are slightly out of tune, and they form a beat frequency.

EDIT: Having analyzed the audio sample, the things you have marked are just some low-frequency artifacts, I think from things like someone touching or shaking the mic or moving around and bumping the guitar body. Adding a 100 Hz high-pass filter mostly gets rid of those:

Sample after 100 Hz high-pass filter
Figure 1. Sample after 100 Hz highpass filtering.

$\endgroup$
  • $\begingroup$ Thanks. I could not recreate this (I forgot which file I used), so see updated question. I also added another subquestion. $\endgroup$ – pavlos163 May 17 '17 at 18:03
0
$\begingroup$

First hypothesis.

Could be an exchange of resonant energy.

Similar to the physics class experiment with two slightly coupled pendulums. You start the first pendulum swinging, then it starts the second swinging while nearly stopping itself, then the reverse happens and the second transfers swing energy back to the first pendulum. Etc. until decay. Whether and how much that happens depend on the ratio of the two pendulum resonant frequencies.

With a guitar, the body is mostly what transfers energy to the air, but it has to first exchange energy with the string. And back and forth.

In reality, a guitar is far more complex, with bunches of mechanically multiply-resonant subcomponents exchanging energy in some complex pattern over time, and then from there to your ear or mic.

$\endgroup$
  • $\begingroup$ Thanks. I could not recreate this (I forgot which file I used), so see updated question. I also added another subquestion. $\endgroup$ – pavlos163 May 17 '17 at 18:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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