The bag of potato chips and its vibrations were monitored from behind a soundproof glass

Aug 5, 2014 20:13 GMT  ·  By
Researchers find a way to recover audio signals from vibrations produced by sound waves
   Researchers find a way to recover audio signals from vibrations produced by sound waves

Legend has it that, if you put a glass up against a wall, you might be able to hear at least some of the things happening or being said on the other side. As it turns out, good old bags of potato chips can work just as well when it comes to spying on people.

In fact, it looks like such snacks might just have what it takes to secure a job as undercover agents, and they have researchers with the Massachusetts Institute of Technology (MIT) to thank for revealing their top secret skills to the world.

Not to beat about the bush, MIT scientists and colleagues at Microsoft and Adobe claim to have recently used a regular bag of potato chips like the one pictured next to this article to recover intelligible speech from vibrations birthed by sound waves.

MIT researchers explain that, when bumping into various objects, sound waves cause them to vibrate. More often than not, the naked eye cannot pick up on these vibrations. High-tech video cameras, however, can pin down such subtle movements.

“When sound hits an object, it causes the object to vibrate. The motion of this vibration creates a very subtle visual signal that’s usually invisible to the naked eye. People didn’t realize that this information was there,” details study lead author Abe Davis.

While experimenting with recovering sounds from vibrations, the scientists behind this research project first employed a high-speed camera able to record somewhere between 2,000 to 6,000 frames per second. This allowed them to reconstruct speech with a high-degree of accuracy.

Interestingly enough, a regular camera that captured merely 60 frames per second was also used. This time, the recovered sounds were not as detailed and accurate. Still, the researchers say that information obtained in this manner could help establish the gender of a speaker.

As detailed on the official website for the Massachusetts Institute of Technology, the audio signal revealed by the bag of chips' vibrations was reconstructed with the help of an algorithm that served to analyze the vibrations as documented with the help of the two cameras.

Just for the record, it must be said that the bag of potato chips the researchers used in their experiments was filmed from a distance of about 15 feet (approximately 4.6 meters), and from behind a soundproof glass. Otherwise put, neither the cameras nor the scientists did any cheating.

Apart from bags of potato chips, the scientists behind this investigation used an aluminum tin foil, the surface of a glass of water, and even the leaves growing on a potted plant to recover and reconstruct audio signals. These object too performed surprisingly well in terms of revealing sound waves.

It is believed that, at some point in the future, this technology will help law enforcement officers and forensics investigator gather evidence they would otherwise have trouble obtaining. Interesting enough, it is believed that the algorithm could also boost the development of a new type of imaging that employs sound waves.

“We’re recovering sounds from objects. That gives us a lot of information about the sound that’s going on around the object, but it also gives us a lot of information about the object itself, because different objects are going to respond to sound in different ways,” says Abe Davis.

“I’m sure there will be applications that nobody will expect. I think the hallmark of good science is when you do something just because it’s cool and then somebody turns around and uses it for something you never imagined. It’s really nice to have this type of creative stuff,” adds researcher Alexei Efros.

Should you have about 5 minutes to spare, check out the video below to learn more about this method of recovering audio signals from the vibrations they cause in the objects sound waves chance to hit.