A question about voiceprinting

Determining whether a particular individual is the speaker in a recorded 
message is pretty routine, especially when the original recording device 
is available for making comparison recordings. MP-3 and some other voice 
compression methods remove much of the "unimportant" detail. I've been 
wondering how much uncertainty that might introduce into voiceprinting. 
Does anyone here know?

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

Jerry Avins wrote:
> Determining whether a particular individual is the speaker in a recorded 
> message is pretty routine, especially when the original recording device 
> is available for making comparison recordings. MP-3 and some other voice 
> compression methods remove much of the "unimportant" detail. I've been 
> wondering how much uncertainty that might introduce into voiceprinting. 
> Does anyone here know?
> 
> Jerry

You might also want to try comp.speech.research . IIRC there have been 
similar questions discussed in last few months.