Also, I am told that humans have a tough time identifying the phase of the frequency components of the sounds they identify. So one could hide date in the phase relationships among the frequency components of, say, recorded speech.
This sounds interesting. How about detection, how will some random phase relationships stand out against normal phase of various frequencies?
I suspect that the unusual phase relationships would stick out, to anyone looking at the speech with,say, an oscilliscope. Eg, a squarewave would look pretty mangled if you shifted its component frequencies by some random amount, even though a human being might not be able to detect the difference between a squarewave, and a component-frequency shifted squarewave. On the other hand, how often does speech over some electronic medeum actually get analyzed this deeply?