May 5, 2025
Trending News

Android vulnerability: Hackers can know your identity and gender

  • December 29, 2022
  • 0

Researchers from five American colleges have discovered a technique for eavesdropping on your conversations and leaking your personal information. This is possible thanks to the sensors found in

Researchers from five American colleges have discovered a technique for eavesdropping on your conversations and leaking your personal information. This is possible thanks to the sensors found in Android devices. A side-channel attack method called EarSpy was developed in collaboration with researchers from five American colleges. The latter does this by capturing the vibrations sent by the speaker. Use of Android smartphone sensors (accelerometer and various gyroscopes). They can be used to get information about the subscriber. It’s like the topic of current conversation once they get a proper analysis.

Researchers from Texas A&M University, Temple University, the University of Dayton, Rutgers University, and the New Jersey Institute of Technology developed a machine learning method to arrive at this conclusion. A side-channel attack like EarSpy wouldn’t be possible without today’s technological advances in component and audio transmission. Trying to use this on the 2016 OnePlus 3T was impossible. But the 2021 OnePlus 9 gave researchers more important information: The speaker of the latest smartphone is much more powerful and accurate.

EarSpy: Android smartphones can be hacked through their sensors
Identification Classifier Data set TP
Evaluation
Rate FP Sensitive Retreat
Sex random forest emo-DB 98.7% 1.3% 98.7% 98.7%
JL-corpus 78.6% 21.7% 78.8% 78.6%
a random subspace emo-DB 84.7% 15.4% 84.7% 84.7%
JL-corpus 79.4% 21.0% 79.8% 79.4%
decision table emo-DB 84.7% 16.7% 84.8% 84.7%
JL-corpus 77.7% 22.5% 77.7% 77.7%
Speaker Forest
JL-corpus 61.5% 13.2% 61.1% 61.5%
a random subspace FSDD 88.7% 5.2% 89.1% 88.7%
JL-corpus 55.7% 15.5% 55.5% 55.7%
decision table FSDD 88.2% 5.4% 88.8% 88.2%
JL-corpus 59.9% 13.7% 59.6% 59.9%
speech random forest FSDD 41.6% 6.8% 41.6% 41.6%
Random Subspace FSDD 39.0% 7.2% 39.1% 39.0%
decision table FSDD 33.3% 8.0% 33.6% 33.3%

EarSpy may sound “promising,” but the data it collects is far from accurate. The scientists were only able to determine the gender of the subscriber with an average accuracy of 88.7%. Even after exercising with a limited amount of audio data. Voice recognition was accurate only 33.3% to 41.6% of the time, while caller ID was accurate only 73.6% of the time. While artificial intelligence is developing rapidly, these rates are rapidly approaching 100%.

If you’re using an Android smartphone, you don’t have to worry yet. Because EarSpy cannot access your conversations or information. The scientists who created this method have our interests in mind. Their research hopes to show how such attacks could one day become widespread. If it was in someone else’s hands. So there is no reason to be afraid.

Source: Port Altele

Leave a Reply

Your email address will not be published. Required fields are marked *

Exit mobile version