Are emitters always receptors? I had a teacher who told us that every emitter, at least electric powered, can also behave as a receptor.
Then he gave us examples of devices producing current :
 - LEDs when put under a light
 - motors when manually used
 - speakers when moved
 - resistors when heated  
This seems a bit simplistic to me, but I lack informations to get rid of this assertion.
Do you have some example of emitters (or receptors !) that would break this theory ?
 A: Your teacher is pointing out that most physical systems obey a time-reversal symmetry. That is, take a process (such as light or sound emission), run the clock backwards, and indeed you'll find that the inverse process happens. So if you set up the system to run in reverse, in this narrow definition the system can act as both an emitter and absorber. Now, there are some situations where time-reversal symmetry is broken, such as with many magnetic systems.  Look up optical isolators, which use magnetism to allow light to propagate only one way through them (not backwards). Put an optical isolator in front of your LED, and the total system will emit light, but not absorb it (the light could be reflected instead). 
As Samuel Weir mentioned in the comments, there is also a complication when you include thermodynamic processes. These include resistors heating up when you run a current through (perhaps causing an incandescent bulb to emit). While each microscopic interaction therein might technically observe time-reversal symmetry, you would never be able to set up the experiment to "run time in reverse". This is because entropy tends to increase with time. If your emitter device starts with a low-entropy state (e.g. a cold resistor carrying a client) and evolves to a high-entropy state (e.g. a hot resistor with a current), you generally won't be able to undo this with an inverse process. This is why you can't take any old resistor, cool it down, and get a voltage drop.
Another example of entropy getting in the way of perfect reciprocity is the LED. An LED will emit light with a color corresponding to its band gap. However, it can absorb light with colors equal to and bluer (that is higher-energy photons) than the band gap. So a red LED can absorb blue light but it won't emit blue light.  That's not to say that it could never emit blue light (the absorption is time-reversal symmetric after all), just that it is exceedingly unlikely to happen because thermodynamically the electrons don't like to stay in the blue states for very long, and they will quickly move down to the red states, losing the excess energy to heat. 
