this post was submitted on 02 Sep 2024
44 points (95.8% liked)

Pulse of Truth

493 readers
116 users here now

Cyber Security news and links to cyber security stories that could make you go hmmm. The content is exactly as it is consumed through RSS feeds and wont be edited (except for the occasional encoding errors).

This community is automagically fed by an instance of Dittybopper.

founded 1 year ago
MODERATORS
 

Comments

you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 1 points 2 months ago* (last edited 2 months ago) (1 children)

It has to. The only way that it wouldn’t trip the indicator is if it was built into the operating system itself or somehow had an exploit to get around OS security protections.

The information is fascinating but by and large should no longer be applicable because the OS has been designed to prevent using the microphone without the users knowledge. An app doesn’t have access to the microphone hardware without going through the OS first. Google could modify the OS to do such a thing, but of course, they have to hide this in the proprietary parts of Android, and the generally open nature of the platform give security researchers quite good access to observe such activity. I’d be surprised such activity would go unnoticed. It seems unlikely.

I think this type of approach might have worked on older OS versions but I don’t see how it could work today in general.

[–] [email protected] 1 points 2 months ago (1 children)
[–] [email protected] 1 points 2 months ago* (last edited 2 months ago) (1 children)

Starting with version 12, the Android operating system introduced a limit of 200Hz to help mitigate such attacks, but as you indicate research shows that some reconstruction may be possible in some scenarios. This is an ongoing area and future mitigations continue to be considered.

From Kaspersky:

In 92% of cases, the accelerometer data made it possible to distinguish one voice from another. In 99% of cases, it was possible to correctly determine gender. Actual speech was recognized with an accuracy of 56% — half of the words could not be reconstructed.

The monitoring application would also need to run in the foreground to access the data on a continuous basis.

Overall it does look like an interesting theoretical concern.

[–] [email protected] 1 points 2 months ago

Here's an exotic conspiracy theory: advertisers are performing sensor fusion / superresolution on many colocated gyrophones to exceed the per-device 200Hz cap. Phone clocks are certainly not aligned to the millisecond, so this would enable them to get a higher time resolution.