Although this method can be used for de-anonymization and fingerprinting, you can't fully reconstruct the screen contents from such a low bandwidth signal.<p>If your private spaces are bugged with microphones or cameras (ie most mobile devices, home audio "assistants", smart TVs, untrusted software etc) expect the owner of the sensor to do the best in their abilities to spy on you.<p>Ultrasonic ad fingerprinting is another interesting way your home spies can communicate with each other.[1]<p>[1] <a href="https://arstechnica.com/tech-policy/2015/11/beware-of-ads-that-use-inaudible-sound-to-link-your-phone-tv-tablet-and-pc/" rel="nofollow">https://arstechnica.com/tech-policy/2015/11/beware-of-ads-th...</a>
> We assume that the attacker knows the content displayed on the attacked monitor, except
for the textual letters. This assumption holds in many cases, for example, when the victim is
filling a form on a known website. We also assume that the font is monospace and sufficiently
large. The requisite size of the text depends on the granularity of the leak, which changes
among different monitors. Another assumption is, again, that the screen is in portrait layout.<p>They go on to state that in their proof of concept they had 3-6 letter word in black against a white background. They recorded audio for 5 seconds, and the letters were 175 px wide using a monospace font.<p>Note that they do mention they expect that any background could work, as long as it is fixed and you can train your model on that background.<p>My point being, this is far from a practical way to read what is on screen.
I enjoy that nobody knows what the Van Eck phreaking[0] is. It works, but to get a good picture the gear is quite large to carry unnoticed.<p>[0]<a href="https://en.wikipedia.org/wiki/Van_Eck_phreaking" rel="nofollow">https://en.wikipedia.org/wiki/Van_Eck_phreaking</a>
IIRC this is what "Tempest for Eliza" was for CRTs<p><a href="http://www.erikyyy.de/tempest/" rel="nofollow">http://www.erikyyy.de/tempest/</a>
In a less paranoic way, there's this related item: <a href="https://news.ycombinator.com/item?id=8862689" rel="nofollow">https://news.ycombinator.com/item?id=8862689</a><p>...which evolved from <a href="https://news.ycombinator.com/item?id=8856829" rel="nofollow">https://news.ycombinator.com/item?id=8856829</a>
It reminds me Tempest (<a href="https://en.m.wikipedia.org/wiki/Tempest_(codename)" rel="nofollow">https://en.m.wikipedia.org/wiki/Tempest_(codename)</a> ).<p>This time they are exploiting information leak that goes to unshielded speakers and can be recorded and replayed easily.<p>Very clever.
I'm reminded of this DEF CON talk from 4 years ago detailing a similar technique: <a href="https://www.youtube.com/watch?v=5N1C3WB8c0o" rel="nofollow">https://www.youtube.com/watch?v=5N1C3WB8c0o</a>
I've noticed this a few years ago, after reading some psychophysics papers, I was generating some pattern images based on what I read, when my monitor started emitting tunes (not by radio emissions), I suspect piezo effect in some capacitor. I instantly realized my monitor could be intercepted accoustically...