> 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.

this seems not to have any practical significance. and recognizing numbers (only 56%) of the time is practically still as good as guessing, and not useful, in the context of spying, without the spoken words