Covert Eavesdropping Through Computer Mice - the "Mic-E-Mouse"

An Anonymous Coward writes:

Covert Eavesdropping through Computer Mice

The abstract from the arXiv paper states:

High-Performance Optical Sensors in Mice expose a critical vulnerability - one where confidential user speech can be leaked. Attackers can exploit these sensors' ever-increasing polling rate and sensitivity to emulate a makeshift microphone and covertly eavesdrop on unsuspecting users. We present an attack vector that capitalizes on acoustic vibrations propagated through the user's work surface, and we show that existing consumer-grade mice can detect these vibrations. However, the collected signal is low-quality and suffers from non-uniform sampling, a non-linear frequency response, and extreme quantization. We introduce Mic-E-Mouse, a pipeline consisting of successive signal processing and machine learning techniques to overcome these challenges and achieve intelligible reconstruction of user speech. We measure Mic-E-Mouse against consumer-grade sensors on the VCTK and AudioMNIST speech datasets, and we achieve an SI-SNR increase of +19, a Speaker-Recognition accuracy of 80% on the automated tests and a WER of 16.79% on the human study

Additional details: Computer mice can eavesdrop on private conversations, researchers discover

High-end computer mice can be used to eavesdrop on the voice conversations of nearby PC users, researchers from the University of California, Irvine, have shown in a new proof-of-concept demonstration.

Given the catchy name 'Mic-E-Mouse' (Microphone-Emulating Mouse), the ingenious technique outlined in Invisible Ears at Your Fingertips: Acoustic Eavesdropping via Mouse Sensors is based on the discovery that some optical mice pick up incredibly small sound vibrations reaching them through the desk surfaces on which they are being used.

These vibrations could then be captured by different types of software on PC, Mac or Linux computers, including non-privileged 'user space' programs such as web browsers or games engines or, failing that, privileged components at OS kernel level.

Although the captured signals were inaudible at first, the team were able to enhance them using Wiener and neural network statistical filtering to boost signal strength relative to noise.

Read more of this story at SoylentNews.