Laser Meager Listener: A Scientific Exploration of Laser-based Speech Eavesdropping in Commercial User Space
Cross Correlation Graphs of Input Audio vs. Reconstructed Audio
* All cross correlation graphs below compare the post-processed reconstructed audio - from vibration data collected in each scenario - to the original source audio that was played during the experiment.
Direct-Contact Scenario:
Paper Cup:
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
Loudspeaker-Same Surface (Normal SPL)
Loudspeaker-Same Surface (Loud SPL)
Plastic Cup:
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
Loudspeaker-Same Surface (Normal SPL)
Loudspeaker-Same Surface (Loud SPL)
Styrofoam Cup:
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
Loudspeaker-Same Surface (Normal SPL)
Loudspeaker-Same Surface (Loud SPL)
Glass-Barrier Scenario:
Paper Cup:
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
Loudspeaker-Same Surface (Normal SPL)
Loudspeaker-Same Surface (Loud SPL)
Plastic Cup:
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
Loudspeaker-Same Surface (Normal SPL)
Loudspeaker-Same Surface (Loud SPL)
Styrofoam Cup:
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
Loudspeaker-Same Surface (Normal SPL)
Loudspeaker-Same Surface (Loud SPL)
Glass-Surface Scenario:
Control (No Speech)
Loudspeaker-Aerial (Normal SPL)
Loudspeaker-Aerial (Loud SPL)
