Type quietly, Big Brother could be listening!

Stuart Corner
Sunday, 18 September 2005 10:13

Business IT - Technology

Researchers at the University of California at Berkeley claim they have been able to determine keystrokes input to a computer keyboard, simply by analysing a sound recording of the typist hitting the keys.

They say they were able to take several 10-minute sound recordings of users typing at a keyboard, feed the audio into a computer, and use an algorithm to recover up to 96 percent of the characters entered. The results of their findings will be presented Nov. 10 at the 12th Association for Computing Machinery Conference on Computer and Communications Security in Alexandria, Virginia.

The researchers claim their findings represent a real threat to information security. "It's a form of acoustical spying that should raise red flags among computer security and privacy experts," said Doug Tygar, UC Berkeley professor of computer science and information management and principal investigator of the study, in a press release announcing the findings.

"The message from this study is that there is no easy escape from this acoustic snooping. The type of keyboard you use doesn't matter, your typing proficiency doesn't matter, and the background noise can be overcome...If we were able to figure this out, it's likely that people with less honourable intentions can - or have - as well."

The technique relies in part on each key sounding slightly different and partly on the statistical characteristics of the English language. "Using statistical learning theory, the computer can categorise the sounds of each key as it's struck and develop a good first guess with an accuracy of 60 percent for characters, and 20 percent for words," said Li Zhuang, a UC Berkeley PhD and lead author of the study. "We then use spelling and grammar checks to refine the results, which increased the character accuracy to 70 percent and the word accuracy to 50 percent. The text is somewhat readable at this point."

The recording is then played back repeatedly in a feedback loop to "train" the computer to increase its accuracy until no significant improvement is seen. In the UC Berkeley experiments, three feedback cycles were often enough to obtain recovery rates of 88 percent for words and 96 percent for characters.

Once the system is trained, recovering the text became more straightforward, even if the text was a password and not an English word. After just 20 attempts, the researchers were able to retrieve 90 percent of five-character passwords, 77 percent of eight-character passwords and 69 percent of 10-character passwords.

The researchers pointed out that they did not use the shift, control, backspace or caps lock keys for their experiments, but describe approaches for training a program to account for those keystrokes as well. The ability to account for use of a computer mouse would be more challenging, the researchers said.

The researchers experimented with multiple users on different keyboards, including so-called "quiet" keyboards, and found that their algorithm was successfully able to predict data. Moreover, recordings were taken in a variety of conditions, such as environments in which music was playing or cell phones were ringing in the background

The full press release, the text used for the experiments and the results after processing are available here.