Researchers at MIT, Microsoft, and Adobe have developed an algorithm that can reconstruct an audio signal from minute vibrations of objects in a video recording, including recovering intelligible speech from the vibrations of a potato-chip bag photographed 15 feet away through soundproof glass.
The researchers also successfully extracted audio from video of aluminum foil, the surface of a glass of water, and the leaves of a potted plant.
I wonder whether the technique might become sensitive enough to capture sound from old silent movies, newsreels, and home movies.
Extracting audio from visual information: Algorithm recovers speech from the vibrations of a potato-chip bag filmed through soundproof glass
Photo: Salt-and-Vinegar, by Gerolsteiner91. Licensed under CC BY-SA 2.5 via Wikimedia Commons
Some venture capitalists at the cutting edge of Internet innovation say they will shun startups requiring fast connections for video, audio, or other services, mindful that the U.S. Federal Communications Commission may let ISPs charge extra fees to major content providers.
MIT Tech Review: Proposed Net Neutrality Rules Already Hurting Innovation – Technology – Boston.com