A conventional microphone contains a diaphragm that vibrates in the presence of sounds waves. These mechanical vibrations are then converted into an electrical signal. In a condenser mic, for example, the diaphragm acts as one plate of a capacitor with the changing capacitance creating the signal.
But diaphragms are not the only things that vibrate in the presence of sound waves – indeed, just about any object will oscillate to a certain extent. Now, researchers at Carnegie Mellon University in the US have created an optical system that can detect vibrations on an object and convert them into an audio signal.
The idea is very simple, a laser is aimed at an object creating a speckle pattern on the surface of the object. The speckle pattern is then imaged using two cameras – one that scans across the pattern, and one that images the entire pattern.
Interfering light
A speckle pattern is caused by the interference of light that is reflected from the microscopic texture on the surface of an object. As a result, the pattern changes as vibrations move across the surface of the objects – and these changes are captured by the cameras and converted into an audio signal using a computer program.
Mark Sheinin, Dorian Chan and colleagues used their system to capture sounds from individual guitars, loudspeakers and even from a bag of Doritos.
This is not the first time that a computer vision system has been used to capture sound, but the team point out that unlike previous systems – which used expensive high-speed cameras – this latest incarnation uses ordinary cameras at a fraction of the cost.
Sound engineers
A paper describing the system won “best paper honourable mention” at the IEEE/CVF Computer Vision and Pattern Recognition Conference, which was held this week in New Orleans. The team says that its system could be used by sound engineers to get a better idea of the sounds emanating from individual instruments. It could also be used to monitor the vibrations of individual machines in manufacturing facilities.
A PDF of the paper can be downloaded here.
For the first time, two young women have been crowned UK Young Scientist of the Year and UK Engineer of the Year in the same year. Connie Gray from Liverpool and Avye Couloute of Surbiton are both 14 and they bagged the science and engineering prizes respectively in The Big Bang UK Young Scientists & Engineers Competition.
Gray’s research involved comparing the structure of birds’ feathers to determine why some birds cannot fly. Her work aims to help with conservation efforts in areas of the world affected by climate change.
Improving air quality
Couloute invented a system that monitors carbon dioxide levels inside buildings. To test her system, she built a scaled-down pavilion and was able to show how the system can automatically improve indoor air quality using ventilation and other techniques.
Congratulations to both winners. And if you would like to be inspired by another young scientific achiever, listen to my interview with 17-year-old Rishab Jain – who is a two-time international science fair winner from the US. I chat with him about his research on the applications of artificial intelligence in medicine.