Picture this, dudes: A program that takes input from the mic and outputs it, reversed, to speakers/headphones/audio-jack-in-your-head, say, 2 samples at a time. Does such a program exist? What programming language should I learn to code one?

Call me stupid, but i don't get it! If it's Real Time then, how could it have reveresed something that doesn't even yet exist??!!!

If you just want to flip the current a previous sample, i don't know if you could do it easily, but i would imagine, the resultant sudio file wouldn't sound to dis-similar.

Kristian

It wouldn't strictly be realtime because reversing a sound sample by sample by sample is impossible. You cannot reverse a single sample. What you could do, though, is pair up as few as two samples, reverse THEM, and then output.



I think you're talking about inversion, not reversal (IE, -4892 becomes 4892). If the difference was perceptible at all, you are right to say it would not be dissimilar.

My gut feeling is that if you reversed sample pairs you would hose up the top octave (11-22 khz) but have little effect below that. So the other poster is right that it would sound pretty much the same.

edit: strictly speaking, if you reversed the samples as they appear in a serial stream you would simply be switching the left and right channels as left and right samples alternate. The prior paragraph is for what I'm sure you meant, reversing samples in the same channel.

DonP, I understand now. It was a difficult problem to wrap my mind around and wishful thinking got the better of me. This idea is TOTALLY wack. Thanks for your input, both of you.