17 Oct
2013
17 Oct
'13
2:11 p.m.
I just came across this Bruce Schneier talk about the big-picture issues of
increasingly powerful technology for communication / surveillance and found
it to be relevant to the discussion here:
https://www.youtube.com/watch?v=h0d_QDgl3gI
On Thu, Oct 17, 2013 at 10:27 AM, Patrik D'haeseleer <patrikd(a)gmail.com>wrote;wrote:
On Thu, Oct 17, 2013 at 6:41 AM, GtwoG PublicOhOne
<g2g-public01(a)att.net>wrote;wrote:
And voiceprint attribution was 99.96% accurate as
of 1962
(published); fast-forward at the speed of Moore's law, draw your own
conclusions, and more details in person.
Note that the accuracy of voiceprint identification has been hugely
exaggerated. That number has been touted primarily by prosecutors, and by
voiceprint identification "experts" who make a nice buck from offering
their services. But just as has been the case with lie detectors, these
claims have not stood up to rigorous scientific validation.
Of course, that doesn't mean that they won't be used against you in a
court of law, regardless of your guilt or innocence! And some
people<http://www.innocenceproject.org/Content/David_Shawn_Pope.php>h… been
falsely convicted to decades in jail or worse on the basis of
this kind of "evidence".
http://sourcesandmethods.blogspot.com/2013/08/is-forensic-speaker-recogniti…
Forensic speaker recognition has many limitations and is currently
inadmissible in federal court as expert testimony. Bonastre et al (2003)
summarize these limitations quite well:
"The term *voiceprint* gives the false impression that voice has
characteristics that are as unique and reliable as fingerprints... this is
absolutely not the case."
The thing about voices is that they are susceptible to a myriad of
external factors such as psychological/emotional state, age, health,
weather... the list goes on. From an application standpoint, the most
prominent of these factors is intentional vocal disguise. There are a
number of things people can intentionally do to their voices to drastically
reduce the ability of machine or human expert to identify their voice
correctly (you would be amazed at how difficult it is - nearly impossible -
to identify a whispered voice). Under these conditions, identification
accuracy falls to 40 - 52 percent (Thompson 1987), 36 percent (Andruski
2007), 26 percent (Clifford 1980).
Patrik
_______________________________________________
sudo-discuss mailing list
sudo-discuss(a)lists.sudoroom.org
http://lists.sudoroom.org/listinfo/sudo-discuss