An experimental project to demonstrate how a user keyboard input may be sniffed through the pattern analysis of the sounds emitted by the keystrokes (based on pyAudioAnalysis).
This field of study is called Acoustic cryptanalysis, (also known as Acoustic Keyboard Eavesdropping) and is a type of side channel attack towards electronic devices.
This project was created as an assignment for the IF754 - Musical Computing class of 2016.1 at CIn/UFPE.
- Rodrigo Alves Vieira <rav2 [at] cin.ufpe.br>
- João Paulo Luna <jpsl2 [at] cin.ufpe.br>
- Edymir Étienne Semedo <eebls [at] cin.ufpe.br>
- EasyClassifier https://sourceforge.net/projects/easyclassifier/
- EchoNest http://the.echonest.com/
- Keyboard sniffing through audio recorded typing patterns http://security.stackexchange.com/questions/23322/keyboard-sniffing-through-audio-recorded-typing-patterns
- Keystroke logging https://en.wikipedia.org/wiki/Keystroke_logging
- Acoustic Snooping on Typed Information https://freedom-to-tinker.com/blog/felten/acoustic-snooping-typed-information/
- All In One Keylogger http://www.relytec.com/help/keylogger_microphone_recording.htm
- Sound Logger Protection https://www.spyshelter.com/sound-logger-protection/
- Fast fourier transform https://en.wikipedia.org/wiki/Fast_Fourier_transform
- Support vector machine https://en.wikipedia.org/wiki/Support_vector_machine
- Dual tone multi frequency signaling https://en.wikipedia.org/wiki/Dual-tone_multi-frequency_signaling
- KNN https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm
© 2016 Rodrigo Alves Vieira, João Paulo Luna & Edymir Étienne Semedo. All Rights Reserved.