Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!EU.net!Germany.EU.net!netmbx.de!zrz.TU-Berlin.DE!cs.tu-berlin.de!fauern!rrze.uni-erlangen.de!hub-n.franken.de!ark.franken.de!ralf
From: ralf@ark.franken.de (Ralf W. Stephan)
Subject: Speaker verification  (was: Simple speech recognition needed)
Message-ID: <1994Sep4.081026.94@ark.franken.de>
Organization: his desk writing an article
Date: Sun, 4 Sep 1994 08:10:26 GMT
References: <344uqf$bog@noc.tor.hookup.net> <347cfe$gf8@cherokee.advtech.uswest.com>
X-Newsreader: TIN [version 1.2 PL2]
Lines: 20

[sorry, this is off topic, but maybe this could start another
 interesting thread...]

Curtis Knittle writes:
> Greg...what you may really be after is a speaker verification 
> system.  I don't think you want a system to just recognize the 
> digits, because then anyone standing around would hear 
> someone elses PIN, and therefore have access to the area.  

But then, someone could record your voice --- and not only when
you say the PIN.  Though this is less likely, it will be a problem.
Now the question:  Is it possible to decide if a certain voice
input comes from a recording, or is synthesized using parameters
that mimick the original speaker?

ralf
--
"It has its good points, but on the whole, TV is way overdone.. There is good
 stuff to watch, but 0.5 hrs a day is an entire week by the end of the year.."
                                                             (Simon Dawson)
