Laser microphones aren't practical at all. In a real application it's unlikely that you'd manage to aim the laser so precisely that its reflection would hit the photocell as the video shows. And that's assuming you'd have a listening post available at an appropriate location. The beam would just reflect off the window and hit something else, and you'd get no signal.

Several problems with this setup:
1. The room would have to be tightly sealed so the SPL would be sufficient to move the window, failing that:
2. The sound source would have to be very near the window
3. The laser source would have to be highly focused and the window would have to be just such an angle to catch the reflection
4. If the window weren't sturdy, the incident beam would reflect far out of the range of the photo-receptor
5. If a fan or furnace (or anything that moved more air than a pair of lungs) were operating in the room, the impact of the voice on the window would be lost in the background noise.

All that said, it's still pretty cool. It's just not very effective.

"Bob, what's that bright red light shining in the corner of the window?"

*engineering

Adding video to channels (Engineering) - requested by oxdottir.