Interesting. If the voltage across the speaker voice coil can be sampled with enough sensitivity at a fast-enough rate, you have an undocumented microphone.
This is true of all speakers
It's true of all dynamic speakers -- the sort with a voice coil and a magnet.
(But not all speakers are dynamic speakers.)
Would this also be true for electrostatic speakers as well? Though would probably would require greater gain/amplification or, potentially the application of some kind of bias voltage for the capacitive diaphragm of the speaker.
Just speculation based on the shared operating principal with condenser microphones
With bias power, I think an electrostatic loudspeaker turns into a condenser microphone (a thing that provides varying capacitance in response to changes in pressure).
I don't think that electrostatic loudspeakers all require bias power, so it's not quite as simple as using a dynamic loudspeaker backwards is.
It is a neat idea, though. A big, flat-panel microphone would be interesting to play with.
You can use a window or any large panel as a microphone without even touching it by observing its vibrations.
You can bounce a laser off it, or even go fully passive using a camera with some sensitivity tricks: I recall a paper that reconstructed a remote conversation by watching a houseplant through a window.
Others who know that better than me and commented but... First time I read that, as a kid, here was I plugging my headphones into the input jack of my parents' soundsystem and, sure enough, it worked as mic (although at as super ultra low volume but I clearly remember it worked).
But not true of all codecs…
Do you think Apple put a hidden microphone in their devices by pure accident?