WOW! M.I.T. Researchers can recreate sound from objects in the room

a reply to: ChaosComplex

This is just flat out wrong and shows a total lack of understanding when it comes to this technology. You said:

Wrong. All you would need is any source of sufficient background noise. Once the noise pollution reaches sufficient saturation, discerning individual wave forms will be next to impossible. It wouldn't matter if you had any amount of speakers between 1 and 134567876543456 and each one utilized an isolation pad. The speaker creates sound by disturbing the air. Disturb the air enough, and it's game over for this tech in it's current state.

Disturb the air enough? This is just silly.

The recreation of sound depends on visual data. They paper says NOTHING about disturbing the air. That's just kooky when you're talking about recreating sound through visual data. You guys are just making it up as you go.

In fact, the speakers were right next to the chips in the experiment and they didn't "disturb the air." That's just nonsense. They wanted to minimize CONTACT VIBRATIONS NOT DISTURB THE AIR VIBRATIONS. That's just kooky talk because this technology recreates sound through visual data.

They didn't want CONTACT VIBRATIONS which would be caused by the speakers and the chips being on the same surface area. Here's the quote from the paper.

The setup for these experiments consisted of an object, a loudspeaker, and the camera, arranged as shown in Figure 4. The loudspeaker was always placed on its own stand separate from the surface holding the object in order to avoid contact vibrations.

This is about CONTACT VIBRATIONS meaning the speakers and the chips would occupy the same surface area. When the chips and the speaker were on different surface areas they carried out the experiment just fine and there wasn't any "DISTURBANCE IN THE AIR."

LOL, you guys just make up this nonsense as you go.

a reply to: neoholographic

The disturbance from airbourme waves and structural waves will have exactly the same effect. In their test setup, the only source is airbourme waves is the speaker. In a real-world environment, well, we've discussed this many, many times already. As nauseum.

a reply to: neoholographic

I think you should rewrite the paper so it makes more sense to us lay people.

originally posted by: neoholographic

This is about CONTACT VIBRATIONS meaning the speakers and the chips would occupy the same surface area. When the chips and the speaker were on different surface areas they carried out the experiment just fine and there wasn't any "DISTURBANCE IN THE AIR."

LOL, you guys just make up this nonsense as you go.

Of course there was a disturbance in the air. How else did the sound waves travel from the speaker to the bag?

originally posted by: neoholographic
a reply to: ChaosComplex

In fact, the speakers were right next to the chips in the experiment and they didn't "disturb the air." That's just nonsense. They wanted to minimize CONTACT VIBRATIONS NOT DISTURB THE AIR VIBRATIONS. That's just kooky talk because this technology recreates sound through visual data.

So what exactly do you think is making the bag vibrate in the first place? How does the sound travel from the speaker to the bag to make it vibrate?

The entire point of a speaker is that it "disturbs the air". If it didn't, it wouldn't work.

The only difference between the sound vibrations and the "contact vibrations" is the medium through which they travels. The source is exactly the same. In the experiment, they only want the vibrations that travel through the air, they don't want the ones that travel through the table.

a reply to: Soylent Green Is People

There is a graphic right above his quote from the paper. It has a symbol for the speakers, then an arrow labeled "Air Pressure (Pa)"

I'm pretty sure sound waves need something to travel through otherwise it would be magic. This is fascinating though.

neoholographic --

I just want to understand your position on this, because it is getting confusing as to what you are trying to say here about this technology. We seem to be going around in circles.

Here's how I see it working:

1. A speaker creates a sound.

2. The sound creates pressure waves through the air (this is how sound travels through air).

3. The chip bag (or "crisp bag" for you non-Americans) is affected by the pressure waves.

4. the chip bag begins to vibrate due to those pressure waves, much like a microphone or speaker diaphragm.

5. The video camera looks at those vibrations, and then analyzes the frequency and amplitude of the vibrations of the bag.

6. The software then electronically recreates the vibrations frequency and amplitude, thus turning it back into sound.

In this way, it is similar to the spy method of bouncing a laser off of a window to hear conversations inside the room on the other side of that window. Using the laser/window method, the laser reads the frequency and amplitude of the window vibrations, similar (albeit different) than the way the video camera and software can read the chip bag vibrations.

Do I have this right? Is that also your understanding of how this "recording sound using a video camera" works?

If not, what is your understanding about how this works?

a reply to: Soylent Green Is People

Here is the graphic from the paper:

I think we all agree that sound produces air pressure. Air disturbance is when someone passes gas. It's a common misconception.

a reply to: Soylent Green Is People

HERE WE GO AGAIN!!!

You guys are debating against something I never said because you're position is just so silly.

I never said AIR PRESSURE has nothing to do with this. This is just another FLAT OUT LIE that you guys are trying to debate against this because you keep making these silly statements that have nothing to do with this technology.

I said, it doesn't create a disturbance in the air that can mask the sound that's being recreated!!

This is just silly nonsense.

The Researchers never said anything about disturbances in the air that can mask the sound that's being recreated. That wouldn't make any sense because the disturbance in the air would have to mask vibrations that are less than 100th of a pixel and are being recreated through visual data.

THEY TALKED ABOUT CONTACT VIBRATIONS NOT ANY DISTURBANCES IN THE AIR THAT CAN MASK THE SOUND THAT'S BEING RECREATED!!

Here's what they said:

The setup for these experiments consisted of an object, a loudspeaker, and the camera, arranged as shown in Figure 4. The loudspeaker was always placed on its own stand separate from the surface holding the object in order to avoid contact vibrations.

So once again, you guys debate against a claim that was never made because your position and claims are absurd and the only thing you can do is make up things to debate.

Show me where I said Air Pressure has nothing to do with this process.

originally posted by: neoholographic
...In fact, the speakers were right next to the chips in the experiment and they didn't "disturb the air." ...

...and there wasn't any "DISTURBANCE IN THE AIR."...

Soundwaves are literally created through disturbing the air.

If you don't understand something so fundamental, there is no way I'm going to be successful through text in this debate.

And with this, I respectfully bow out of this circular conversation.

Best of luck to you in future studies.

originally posted by: neoholographic
a reply to: Soylent Green Is People

I said, it doesn't create a disturbance in the air that can mask the sound that's being recreated!!

Yes it does. How else do you think the sound is being transmitted? Magic?

The Researchers never said anything about disturbances in the air that can mask the sound that's being created.

That's because research papers presume a certain level of understanding that you seem to be lacking.

That wouldn't make any sense because the disturbance in the air would have to mask vibrations that are less than 100th of a pixel and are being recreated through visual data.

They are measuring "disturbance in the air"... it's called "sound"!! A mechanical wave travelling though the air. Any "disturbance in the air" that is NOT the sound you wish to record is considered noise and will mask the signal. It is this "disturbance in the air" that is causing the bag to vibrate in the first place. Haven't we already been through this multiple times already?

THEY TALKED ABOUT CONTACT VIBRATIONS NOT ANY DISTURBANCES IN THE AIR THAT CAN MASK THE SOUND THAT'S BEING RECREATED!!

That's because in their experimental setup there is only ONE source of sound travelling through the air. In a real-world setup there will most likely be MULTIPLE sources of sound travelling through the air, resulting in.... I'll leave you to fill the gap here.

Here's what they said:

The setup for these experiments consisted of an object, a loudspeaker, and the camera, arranged as shown in Figure 4. The loudspeaker was always placed on its own stand separate from the surface holding the object in order to avoid contact vibrations.

Because the setup is evidently sensitive enough that even flanking transmission from the speaker to the object (i.e. travelling through the structure) is enough to affect the results, hence the decoupling.

So once again, you guys debate against a claim that was never made because your position and claims are absurd and the only thing you can do is make up things to debate.

Show where I said Air Pressure has nothing to do with this process.

You don't even understand your own arguments, let alone other people's arguments.

a reply to: ChaosComplex

Nobody said soundwaves weren't created by disturbing the air. Again, this is just silliness because the position you support has nothing to do with this technology. I said that a disturbance in the air wouldn't cause vibrations that would mask the recreation of sound.

This is what I said in direct response to your asinine diatribe:

Disturb the air enough? This is just silly.

No, disturbance in the air wouldn't disturb the air enough to mask the sound that's being created.

I then said:

They wanted to minimize CONTACT VIBRATIONS NOT DISTURB THE AIR VIBRATIONS. That's just kooky talk because this technology recreates sound through visual data.

You guys have a bad habit of making up stuff and then debating against those things you're making up.

I never said soundwaves don't create a disturbance in the air. I said the vibrations from this disturbance can't mask the sound that's being recreated and this is why the Researchers talked about CONTACT VIBRATIONS and made sure the speaker and the chips were not on the same surface.

a reply to: neoholographic

If someone throws a rock in the water, would you consider the waves created a disturbance in the water? The waves that hit an object would cause the object to move. The movement of the object would be caused by the change in water pressure. I think this is a pretty basic idea. Do you disagree? Do you have a different understanding?

originally posted by: neoholographic
a reply to: ChaosComplex

No, disturbance in the air wouldn't disturb the air enough to mask the sound that's being created.

Yest it would. I'm sorry you're still not grasping this.

I then said:

They wanted to minimize CONTACT VIBRATIONS NOT DISTURB THE AIR VIBRATIONS. That's just kooky talk because this technology recreates sound through visual data.


I never said soundwaves don't create a disturbance in the air. I said the vibrations from this disturbance can't mask the sound that's being recreated and this is why the Researchers talked about CONTACT VIBRATIONS and made sure the speaker and the chips were not on the same surface.

Wrong again. The "contact vibrations" are but another source of noise. FYI:

Acoustic transmission in building design refers to a number of processes by which sound can be transferred from one part of a building to another. Typically these are:

1. Airborne transmission - a noise source in one room sends air pressure waves which induce vibration to one side of a wall or element of structure setting it moving such that the other face of the wall vibrates in an adjacent room. Structural isolation therefore becomes an important consideration in the acoustic design of buildings. Highly sensitive areas of buildings, for example recording studios, may be almost entirely isolated from the rest of a structure by constructing the studios as effective boxes supported by springs. Air tightness also becomes an important control technique. A tightly sealed door might have reasonable sound reduction properties, but if it is left open only a few millimeters its effectiveness is reduced to practically nothing. The most important acoustic control method is adding mass into the structure, such as a heavy dividing wall, which will usually reduce airborne sound transmission better than a light one.

2. Impact transmission - a noise source in one room results from an impact of an object onto a separating surface, such as a floor and transmits the sound to an adjacent room. A typical example would be the sound of footsteps in a room being heard in a room below. Acoustic control measures usually include attempts to isolate the source of the impact, or cushioning it. For example carpets will perform significantly better than hard floors.

3. Flanking transmission - a more complex form of noise transmission, where the resultant vibrations from a noise source are transmitted to other rooms of the building usually by elements of structure within the building. For example, in a steel framed building, once the frame itself is set into motion the effective transmission can be pronounced.

en.wikipedia.org...

They have minimized 1) by only having one audio source in their experimental setup. They have minimized 2) and particularly 3) by decoupling the speaker and object from the surface.

And around we go...

a reply to: ZetaRediculian

You guys have stepped in it once again LOL!!

Again:

A disturbance in the air will not create vibrations that can mask the sound that's being recreated unless there's contact vibrations. If you put the bag of chips and speakers and separate surfaces then the disturbance in the air wouldn't be enough to mask the sound that's being recreated.

This is why the researchers SEPARATED THE CHIPS AND THE SPEAKER because they didn't want them on the same surface to avoid contact vibrations. They carried out the experiment just fine and there wasn't any disturbance in the air that could mask the sound that was being recreated.

The setup for these experiments consisted of an object, a loudspeaker, and the camera, arranged as shown in Figure 4. The loudspeaker was always placed on its own stand separate from the surface holding the object in order to avoid contact vibrations.

THIS JUST TAKES A LITTLE COMMON SENSE TO UNDERSTAND!!

It's obvious that you guys want to debate against things that I never said and this is because your positions are so asinine!!

a reply to: neoholographic

A disturbance in the air will not create vibrations that can mask the sound that's being recreated unless there's contact vibrations.

Wrong again. See my previous post.

Explain this:

Replace the speaker with two people talking. How does the equipment pick out Person A's voice over Person B's?

a reply to: GetHyped

Again, debating a claim I never made. You said:

Wrong again. The "contact vibrations" are but another source of noise.

SHOW ME WHERE I SAID CONTACT VIBRATIONS WASN'T ANOTHER SOURCE OF NOISE!!

Please stop it you guys, you're looking very ignorant here.

TIME AFTER TIME YOU DEBATE AGAINST THINGS THAT I NEVER SAID.

If we're going to debate the issue then please debate against things that I actually say. I have had to write post after post like this because you guys can't debate the things I actually say. You have to make stuff up and then debate against the things you're making up.



I said once you put the speaker and the chips on separate surfaces you avoid contact vibrations and the disturbance in the air isn't enough to mask the sounds that's being created and once they avoided contact vibrations they carried out the experiment.

a reply to: neoholographic

You said:

A disturbance in the air will not create vibrations that can mask the sound that's being recreated unless there's contact vibrations.

Which is flat out wrong. You're not even being consistent with your misunderstanding. Again, the answer to this question should reveal the crux of your misunderstanding:

Replace the speaker with two people talking. How does the equipment pick out Person A's voice over Person B's?

a reply to: neoholographic

I am trying to understand what you ARE saying which quite difficult.

I think your position is that this process is somehow immune to noise. Can you explain what you are actually trying to say without all the usual flamboyancy?

you just seem to be going around in circles. First there is no unwanted vibrations, then the paper is shown, then the paper has nothing to do with anything, then you quote the paper, then you say there is no disturbance in the air, then you are shown the graphic from the paper, then there is disturbance in the air.

so can you very calmly say what you mean?

a reply to: GetHyped

It's obvious you don't understand the technology. You said:

Replace the speaker with two people talking. How does the equipment pick out Person A's voice over Person B's?

First of why would the equipment need to pick out person A's voice over person B's voice when the sound is being recreated as a whole through the algorithm?

All you have to do is read about the technology.

The researchers developed an algorithm that combines the output of the filters to infer the motions of an object as a whole when it’s struck by sound waves. Different edges of the object may be moving in different directions, so the algorithm first aligns all the measurements so that they won’t cancel each other out. And it gives greater weight to measurements made at very distinct edges — clear boundaries between different color values.

In fact they could even make out the gender of different people speaking with a camera with a lower fps.

In other experiments, however, they used an ordinary digital camera. Because of a quirk in the design of most cameras’ sensors, the researchers were able to infer information about high-frequency vibrations even from video recorded at a standard 60 frames per second.

While this audio reconstruction wasn’t as faithful as that with the high-speed camera, it may still be good enough to identify the gender of a speaker in a room; the number of speakers; and even, given accurate enough information about the acoustic properties of speakers’ voices, their identities.

Even with a camera with 60 fps, they were able to identify the gender of the speaker, the number of speakers and give accurate information about the acoustic properties of the speakers voices.

If you just read about the technology, you would understand this. And like I said, they did this with a camera at 60 fps vs. a camera that's over 5,000 fps.

It may have been harder to extract this information if there were CONTACT VIBRATIONS. THIS IS WHY WHEN THEY CARRIED OUT ONE OF THE EXPERIMENTS THEY SAID THIS.

The setup for these experiments consisted of an object, a loudspeaker, and the camera, arranged as shown in Figure 4. The loudspeaker was always placed on its own stand separate from the surface holding the object in order to avoid contact vibrations.