[18471] in cryptography@c2.net mail archive
Record a keyboard, reconstruct what was typed.
daemon@ATHENA.MIT.EDU (Perry E. Metzger)
Thu Sep 15 10:35:14 2005
X-Original-To: cryptography@metzdowd.com
X-Original-To: cryptography@metzdowd.com
To: cryptography@metzdowd.com
From: "Perry E. Metzger" <perry@piermont.com>
Date: Thu, 15 Sep 2005 10:12:39 -0400
Interesting new paper:
http://www.cs.berkeley.edu/~tygar/papers/Keyboard_Acoustic_Emanations_Revisited/preprint.pdf
We examine the problem of keyboard acoustic emanations. We
present a novel attack taking as input a 10-minute sound recording
of a user typing English text using a keyboard, and then recovering
up to 96% of typed characters. There is no need for a labeled
training recording. Moreover the recognizer bootstrapped this way
can even recognize random text such as passwords: In our experiments,
90% of 5-character random passwords using only letters can
be generated in fewer than 20 attempts by an adversary; 80% of 10-
character passwords can be generated in fewer than 75 attempts.
Our attack uses the statistical constraints of the underlying content,
English language, to reconstruct text from sound recordings
without any labeled training data. The attack uses a combination
of standard machine learning and speech recognition techniques,
including cepstrum features, Hidden Markov Models, linear classi-
fication, and feedback-based incremental learning.
Perry
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo@metzdowd.com
