Motion sensors available on current smartphone platforms may be sensitive to speech signals. From security and privacy perspective, this raises a serious concern regarding sensitive speech reconstruction, and speaker or gender identification by a malicious application having unrestricted access to motion sensor readings, without using the microphone.
We closely inspect the effect of speech on smartphone motion sensors, in particular, gyroscope and accelerometer. Our aim is to determine the level of threat posed by motion sensors to speech privacy in different scenarios that resemble real life situations. We examine the possibility of surface conduction of vibrations from the speaker to the motion sensors as an alternative pathway for inducing an effect on the motion sensors. Our scenarios involve using laptop speakers and smartphone speakers as possible sources of speech that may be compromised in the light of sensitivity of motion sensors to sound waves or vibrations. We also look into live human speech (both male and female) to determine if it is powerful enough to invoke a response in the motion sensors.
- S Abhishek Anand (PhD candidate)