[s_q_b]

You can also extrapolate a fairly decent image from the microphone, sort of like a make-shift SONAR.

[usefulcat]

From multiple microphones, perhaps, but from a single microphone? Seems unlikely.

[s_q_b]

You have an array of microphones in your phone alone, not to mention the speakers, which are just low-fi mics. But in any case, you can do it with one microphone.

E.g. http://fab.cba.mit.edu/classes/862.13/students/brandon/index...

http://www.extremetech.com/computing/128735-microsoft-create...

[fyrabanks]

That first technique A) is very audibly noticeable B) requires you to put the laptop on a chair and wheel it around the room to make up for the fact that you only have a single microphone.

The second technique basically only understands gestures.

I wouldn't be at all worried about this in the wild...

[s_q_b]

Someone actually did it to China's PLA Unit 61398 using a technique similar to this: http://www.gizmag.com/microphone-room-mapping-epfl/27985/, so I'd say it's worth worrying about in the wild.

[dTal]

Its unlikely the speakers can be used as microphones, as they will be wired up to D/A converters and not D/A converters.

[logfromblammo]

You can produce a depth map from a single lens camera by manipulating the flash across multiple captured images. Similar results could be achieved with a single microphone if the device also has speakers capable of producing sound at certain frequencies.