We also investigate why different keys produceĭifferent sounds and provide hints for the design of homophonic Our attack employs a neural network to recognize

We show that PC keyboards, notebook keyboards, telephone and ATM padsĪre vulnerable to attacks based on differentiating the sound emanatedīy different keys. The following is the abstract of that paper: In 2004 Dmitri Asonov and Rakesh Agrawal from IBM published a paper (pdf) that describes such an attack. This is known as 'Acoustic Keyboard Eavesdropping'.