Hello,

I'm looking to record a mob informant type video where the subject is interviewed and appears as a silhouette. See an example here:

Does anyone have any tips on how to do this with simple equipment at home? (No I am not a mob informant. I want to use this as kind of a marketing gimmick for a website that I am setting up).

Thanks in advance for any pointers!

Not my area of photography BUT as a guess ------ light only behind the subject, no light in front of the subject, meter from the brightest point of the lit wall behind with a small aperture (say f16).

Thanks. Will give it a try!

Don't forget to adjust the sound of the speakers voice.

Don't forget to adjust the sound of the speakers voice.

Thanks, any idea where I can get a device in Bangkok that does that? (I don't want to use a software solution as surely it can be reversed).

Don't forget to adjust the sound of the speakers voice.

Thanks, any idea where I can get a device in Bangkok that does that? (I don't want to use a software solution as surely it can be reversed).

No idea expect for maybe using a tin can with a microphone placed inside of it or have someone not involved, read a script. Just had an idea... How about using some text to speech software. I have never used them so I cannot comment on the quality or which one is best. I found this free one http://www.naturalreaders.com/index.htm and there are others. Maybe that could be a solution.

If it's a marketing gimmick, what's wrong if someone figures out the software solution?

And in either case, choose a random destructive waveform editor of your choice, pitch up an octave, add a light flange/chorusing/reverb/whatever you like with your own settings (just tweak the knobs/sliders a bit), then repitch down to original pitch. That should destroy enough resolution/recognizability.

Add a bitcrusher if you really want to destroy some information.

Incredibly hard to reverse if no one knows what you used (every software does it differently/uses different algorithms, and random tweaks in the right way are unlikely to be replicated).

To get technical, someone could get pretty close if they knew the original voice. Which I could see if perhaps someone is checking if Person X is actually Mr. Bob Smith from down the street. But if it's a marketing ploy, why go through the trouble? Just downpitch and bitcrush.