University of California, Berkeley researchers have developed a system that analyzes audio recordings of keyboard clicks to reconstruct an accurate transcript of what has been typed on a computer. Their method relies on the fact that each key produces a characteristic click, shaped by its position on the keyboard, the vigor and hand position of the typist, and the type of keyboard.
The system, developed by Berkeley's Doug Tygar, uses methods developed for speech-recognition software to group together all the similar-sounding keystrokes in a recording, generating an alphabet of clicks. The software assigns each click a letter based on its frequency, then tests the message created by this assignment using statistical models of the English language. In order to apply the language model, at least five minutes of recorded typing must be in standard English. Once that requirement is met, the program can decode anything from epic prose to randomized, 10-character passwords.
From The Economist
View Full Article – May Require Free Registration
Abstracts Copyright © 2010 Information Inc., Bethesda, Maryland, USA
No entries found