[CodeMichael]

Correct, last article I recall reading about deciphering VBR from packet size alone was something in the neighborhood of 50% success rate.

[Y_Y]

Then why not compress really efficiently by just transmitting packet sizes?

[c22]

Because 50% of them won't be understood?

[e12e]

On the other hand, if you could do it, you'd probably have invented a convoluted speech-to-text (where text is a index into a dictionary of words). Note that you would also likely lose things like inflection, voice, accent etc - so while it might work as a texting system with voice input - it would be a poor substitute for voice chat..