>On a side note there is a known vulnerability in how compressed & encrypted voice data is transmitted that makes it reveal the contents, and that is by analyzing the statistical distribution of data (even encrypted) from that it is possible to deduce the message content -- http://www.cs.jhu.edu/~cwright/oakland08.pdf so, make sure to play relatively loud music in the background

It seems like the better solution would be to just use fixed bitrate compression.