Live data from Hacker News

SPEAKE(a)R: Turn Speakers to Microphones for Fun and Profit [pdf] (2017)

usenix.org

21–30 of 77 posts

Re: SPEAKE(a)R: Turn Speakers to Microphones for Fun and Profit [pdf] (2017)

#21
post #5

Earlier quoted context omitted.

This shouldn't be downvoted. Transducers being reversible is a neat and non-obvious thing.

Motors can be dynamos too

and many LEDs are weak photo-diodes, i.e. you get weak current when you shine a light to them.

Re: SPEAKE(a)R: Turn Speakers to Microphones for Fun and Profit [pdf] (2017)

#26
post #19

If this or an accelerometer based recording is what Meta uses to eavesdrop on in-person talk then color me pink

It's pretty unlikely that Meta is actually eavesdropping on your conversations, because it'd be immediately obvious from battery usage. The ability to turn speakers into microphones doesn't help if the speakers aren't actually connected to an ADC, and both of the modern smartphone OSes limit you to on the order of hundreds of samples per second, so it's rather difficult to get anything sensible without either doing a bunch of local analysis or exfiltrating it, both of which would be visible.

Re: SPEAKE(a)R: Turn Speakers to Microphones for Fun and Profit [pdf] (2017)

#27
post #10

A magnet in a coil operates both ways, this is non intuitive but perfectly sound. Not sure if it's mentioned in the article but microphones can be speakers too...

Not sure if it's mentioned in the article but microphones can be speakers too... Only dynamic mics, which are relatively rare and seldom encountered without an attached preamp. The vast majority of mics for PCs are condensers and electrets. Anything can be a speaker, briefly and only once, if you apply enough voltage to it...

Huh? The standard stage mic, the Shure SM58, certainly is dynamic and has no preamp.

But you probsbly think about smaller form mics like found on headsets (Electrets).

Re: SPEAKE(a)R: Turn Speakers to Microphones for Fun and Profit [pdf] (2017)

#30
post #19

If this or an accelerometer based recording is what Meta uses to eavesdrop on in-person talk then color me pink

It's pretty unlikely that Meta is actually eavesdropping on your conversations, because it'd be immediately obvious from battery usage. The ability to turn speakers into microphones doesn't help if the speakers aren't actually connected to an ADC, and both of the modern smartphone OSes limit you to on the order of hundreds of samples per second, so it's rather difficult to get anything sensible without either doing a…

It can be done with neural networks [1]. Also, speech doesn’t need much bandwidth to be intelligible. You would need control of the analog filter between the accelerometer and the ADC. With 250/s acceleration samples you can reconstruct a signal of a bandwidth of more than 100 Hz anywhere in the spectrum. That is called undersampling.

[1] https://dl.acm.org/doi/abs/10.1145/3478102

Post reply on HN