EarSpy: Spying On Phone Calls Via Ear Speaker Vibrations Captured By Accelerometer

An anonymous reader quotes a report from SecurityWeek: As smartphone manufacturers are improving the ear speakers in their devices, it can become easier for malicious actors to leverage a particular side-channel for eavesdropping on a targeted user's conversations, according to a team of researchers from several universities in the United States. The attack method, named EarSpy, is described in a paper published just before Christmas by researchers from Texas A&M University, Temple University, New Jersey Institute of Technology, Rutgers University, and the University of Dayton. EarSpy relies on the phone's ear speaker -- the speaker at the top of the device that is used when the phone is held to the ear -- and the device's built-in accelerometer for capturing the tiny vibrations generated by the speaker. The researchers discovered that attacks such as EarSpy are becoming increasingly feasible due to the improvements made by smartphone manufacturers to ear speakers. They conducted tests on the OnePlus 7T and the OnePlus 9 smartphones -- both running Android -- and found that significantly more data can be captured by the accelerometer from the ear speaker due to the stereo speakers present in these newer models compared to the older model OnePlus phones, which did not have stereo speakers. The experiments conducted by the academic researchers analyzed the reverberation effect of ear speakers on the accelerometer by extracting time-frequency domain features and spectrograms. The analysis focused on gender recognition, speaker recognition, and speech recognition. In the gender recognition test, whose goal is to determine whether the target is male or female, the EarSpy attack had a 98% accuracy. The accuracy was nearly as high, at 92%, for detecting the speaker's identity. When it comes to actual speech, the accuracy was up to 56% for capturing digits spoken in a phone call. "[This] accuracy still exhibits five times greater accuracy than a random guess, which implies that vibration due to the ear speaker induced a reasonable amount of distinguishable impact on accelerometer data," the researchers said.

Read more of this story at Slashdot.