Massachusetts Institute of Technology, MIT, has found a new method to record audio using a high-speed videocamera and computer analysis. The videocamera registers the small vibrations of sound on a object’s surface and the computer analyses these images using additional Signal-To-Noise metrics.
Check the following sound examples.
Sound recovered from video (via a bag of chips!):
MIT also made this interesting video to explain this new technique:
In photography a lot of innovation is going on. In videography as well. People are using high resolution cameras, using RAW data which can be further processed. Color grading. Adding noise. Using for example the unbelievable sensitive Sony A7s camera (shoot a video at night and it will look like it’s daytime! – check for example this post by the fantastic videomaker Philip Bloom).
Maybe in the future we will be able to use a videocamera for recording both sound and images. Using software which will analyse the vibrations on the shirt of the person who’s being interviewed, or the vibrations of the hairs on the cheek for creating that perfect sound. Software processing is key in all this. In the future we will we able to capture and recreate sound in totally different ways that’s for sure. We might find new ways to filter out noise and new ways to capture sound over long distances.
Are you ready for a paradigm shift? I am!
For more information about the Visual Microphone see this page at MIT.