2
$\begingroup$

Not sure if this is the right place to ask this but here goes...

I've been watching Daredevil and reading a few StackExchange questions here and there and got to wondering...

If someone/something with echolocation powers was listening to an audio recording, like music, or "listening" to a movie, would they be able to sense the performers/studio technicians that were in the studio when the recording happened? Could this be a way to get 3d "video" of the past?

The two main problems I see are:

  1. Echolocation usually relies on being able to send your own signals, like dolphin clicks or bat screeches. You can't project your sounds into the world of the recording.
  2. Bitrate of the recording would limit the amount of info to be gained. Likewise anything with only a single channel of audio would greatly reduce the ability to extract information from it. The more channels the better

How much detail could possibly be extracted from audio recordings about the environment in which they were recorded?

$\endgroup$
4
$\begingroup$

Typically echolocation skills are very tailored to one's individual body, and that makes it hard to use on movies. Movies are typically not recorded with the kind of microphone setups needed to capture the behaviors properly. One might be able to cleverly reverse engineer some mic positions, but to use those skills inuitively, one needs to have recorded the original with something like this.

Binaural recorder

Tools like the 3Dio are designed to match the acoustic behavior of the average human head/ears. However, the results are typically only useful for headphones. If you intend your movie to be listened to over speakers, binaural recording devices like this are really difficult because they don't give much freedom as to what appears in each channel.

Even if you used a binaural recording device, you still would have trouble. See those squggly folds in your ear? They're not just for decoration. One of their effects is to "color" the sound differently, depending on what direction the sound comes from. The pinna, as it is called, along with your shoulders makes things sound different depending on which direction it came from. Your binaural recording would have to have a mold of your ears before this effect would be accounted for.

Think its a joke? Try this experiment with someone. I've had it work for me, and I'd love to have others try and report on the success.

  • Find two spoons that you can clack together. Make sure you do not clack them together in the presence of your test subject. The goal is for them to do the first test without knowing what they're "supposed" to sound like.
  • Have your test subject close their eyes.
  • Clack the spoons together, and have them reach for where they believe the spoons actually are. Do this several times. If your results are like mine, you'll find they tend to get the horizontal direction rather correct, but they're often off in the vertical direction. This is because their ears can use timing differences between left and right ears to figure out left-to-right, but you can't do the same for up and down.
  • Have the test subject open their eyes.
  • Clack the spoons together in a few locations. They now get to hear the sound and see the correct angular information. They know their ears well enough that they can reverse the "coloring" of the sound and learn what the true sound of the spoons are.
  • Have them close their eyes again, and repeat the test. They will likely reach directly for the spoons every time, because now they know what the spoons are "supposed" to sound like, and they can compare that to the colored sound they hear to figure out where the spoons are.

I've done this experiment, with uncanny results. It indeed does appear that one can hear with one's eyes! It just shows how much more information goes into our heads than we think does!

$\endgroup$
  • $\begingroup$ So a single sound heard directly uses timing and volume differences to get one axis, and coloring to get the other. That doesn't have anything to do with echolocation, using timing of echos to infir distance. I suppose it applies to locating the surface doing the reflecting. $\endgroup$ – JDługosz Apr 7 '16 at 16:05
  • $\begingroup$ @JDługosz You're right that its not echolocation. I was using the more general sense of being able to passively detect objects using sound because, as Nacht pointed out, the inability to make noise limits your ability to use echos. $\endgroup$ – Cort Ammon - Reinstate Monica Apr 7 '16 at 18:15
  • $\begingroup$ echolocation would require both distance and position. You guys all have good answers but I guess I learned the most from this one. $\endgroup$ – Nacht - Reinstate Monica Apr 11 '16 at 4:06
3
$\begingroup$

I would say extremely little chance (although distance is more plausible). Recording studios (or anywhere that movies/musicians etc record audio) are setup in a way to limit any and all external sounds or are corrected via software that removes all excess noises. While recording, artists wear high end headphones to play the music in their ears. Movies, most background sounds are added in via software after the picture is recorded as filming in the middle of hollywood (or elsewhere) can be very noisy compared to what the director wants to contray. The most you could determine would be possibly the size of the room/"shapeish" and even then that is an incredibly small chance with audio quality being cut down during the formatting/copy/burn stage. Yes Echolocation usually references your own sounds, but with enough listening(daredevil powers) you may possibly be able to tell when his/her head is turned to give an estimate of shape/distance)

You would not get a 3d "picture" of the environment, but you may be able to get an idea of distances from the "focal point"

$\endgroup$
3
$\begingroup$

I recall this being done with multiple recordings made that record the shots fired in the JFK assassination. So yes, it can be done in extreme cases like a gunshot which gives a clear impuse-response in the recording.

Normal recordings express the sound of the room through reverberation. This is often carefully controlled to sound "right", like a hall of the desired size. The casual listener can tell if the room was small or large, etc.

The drummer's sound often gets picked up by all the other tracks, despite the best efforts to isolate it. If the drum sound can be distinguished from other sounds, it's certainly possible to determine the distances of the various reflections and perhaps figure out more about the room.

As far as listening to a finished product, no. The stray sounds are purposfully removed, and various tracks are mixed together with different levels, other modifications, and timing offsets.

$\endgroup$

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.