« | Home | Recent Comments | Categories | »

Boneh Publications: Gyrophone: Recognizing Speech From Gyroscope Signals

Posted on August 23rd, 2014 at 23:42 by John Sinteur in category: Apple, Google -- Write a comment


We show that the MEMS gyroscopes found on modern smart phones are sufficiently sensitive to measure acoustic signals in the vicinity of the phone. The resulting signals contain only very low-frequency information (<200Hz). Nevertheless we show, using signal processing and machine learning, that this information is sufficient to identify speaker information and even parse speech. Since iOS and Android require no special permissions to access the gyro, our results show that apps and active web content that cannot access the microphone can nevertheless eavesdrop on speech in the vicinity of the phone.

  1. Angela Merkel’s phone just got dropped down an elevator shaft, again.

  2. Bit iOS apps don’t get to do much in the background, so is this feasible on iOS?

  3. Correct. That’s why all those fitness apps use the gps+accelerometer instead.

previous post: Fungus Deadly to AIDS Patients Found to Grow on Trees

next post: Massive Tree Pollen Explosion