Trayvon Martin, Vigilante Justice

[TheL8Elvis]

That's not the threshold in science.

OK - what is the threshold in science, and when and how did easy voice software meet said threshold ?

 

[RandFan]

OK - what is the threshold in science, and when and how did easy voice software meet said threshold ?

Depends on the field but it's no more for me to say what is the threshold than it is for you to say what isn't the threshold. It's not my job to tell DNA forensic experts what is and isn't the threshold anymore than it is for you tell audio experts.

 

[Dancing David]

I might agree with you if I thought the certainty in DNA testing was equivalent to voice recognition (with limited data in this case).

As an a side, it is common in DNA evidence to give population and prevalence statistics, to show the likelyhood of a random false positive.

 

[Dancing David]

You do realize that Zimmerman is not shown with any kind of bandage or wound dressing for an injury supposedly severe enough to require stitches, don't you?

Maybe you can be the first person to account for that.

And you have this from first hand accounts and photos, how?

Oh from a video on the internet.

 

[WildCat]

For the millionth time.... positively excluding someone is not the same as failing to make a match (for example due to poor conditions). Poor audio would affect whether a source voice could be eliminated, if the audio was too poor, the result would come back as inconclusive, not as elimination.

Uh, no, the test concludes no such thing, only the examiner can determine that. The software can't tell whether or not the audio sample is a human voice or a barking dog.

 

[tyr_13]

For the millionth time.... positively excluding someone is not the same as failing to make a match (for example due to poor conditions). Poor audio would affect whether a source voice could be eliminated, if the audio was too poor, the result would come back as inconclusive, not as elimination. It may well be that the media have conflated these options (and in my opinion they have really muddied the waters due to their poor understanding of what the experts are saying), but it is unlikely that the experts have. From what I have heard from the experts they were saying that they had enough to positively exclude Zimmerman with reasonable certainty.

As usual, all of the above is based on the assumption the techniques having the validity they are claimed to have. If they turn out not to be valid and reliable tests, then all bets are off. I have yet to hear of any other scientists claiming there is a problem with them, though.

In that case change 'exclude' Martin to 'positively exclude' Martin. It would still be a way to falsify the exclusion of Zimmerman.

I absolutely understand the objection. Either the science is reliable with certain error bars or it is not. If I can reliably exclude Zimmerman then I can exclude Zimmerman. You are simply voicing your concerns about the science and expressing what would convince you. Science does not work that way. DNA does not carry such a requirement in most minds. Some people don't trust DNA. They could demand the same as you do.

That's not the threshold in science.

DNA already overcame that threshold for certain situations. New uses of DNA had to overcome it each time it was applied in a situation that might have confounding factors. Testing the implications of error bars, attempting to falsify a hypothesis, is a threshold in science.

 

[leftysergeant]

Am I missing something here? Only one suspect was apprehended (that he knows of), but he knows that all of the 8 burglaries were by young black males? there were eye-witnesses to all 8 burglaries? Some crap burglars they must have there - and by extension some crap police work....

Maybe it's because the only two people, as far as we can tell, on the Neighborhood Watch are violent sociopaths who are just looking for a way to boost their highly suspect virility and have no bloody clue what they are doing. If this is the same Frank Taaffe, he is a sorry sack of crap to whom no person should turn for any kind of assistsance or advice.

 

[Professor Yaffle]

Uh, no, the test concludes no such thing, only the examiner can determine that. The software can't tell whether or not the audio sample is a human voice or a barking dog.

Well yes, neither can a DNA test tell the species of the animal the DNA came from.... Your point?

 

[leftysergeant]

Why in the hell would it occur to Owen to do that since his software doesn't require it, right?

The basic principles of forensic science require it.

 

[RandFan]

DNA already overcame that threshold for certain situations. New uses of DNA had to overcome it each time it was applied in a situation that might have confounding factors. Testing the implications of error bars, attempting to falsify a hypothesis, is a threshold in science.

What objective basis do you have to suppose that voice printing is so unreliable as to be unable to exclude GZ as it is now? Or are you speculating based on your understanding of the technology.

 

[Cylinder]

The basic principles of forensic science require it.

Yea. I'm coming to that conclusion as well.

 

[WildCat]

Well yes, neither can a DNA test tell the species of the animal the DNA came from.... Your point?

Standards have to be met before the sample is even input to the software. And this particular sample violates just about every published scientific standard.

Unless there's a published paper detailing how a poor quality audio sample of a short scream for help can be used in comparing to a sample of a spoken voice in which the word "help" is never uttered? If so, feel free to present it. No one has so far in this thread...

 

[Kestrel]

The gate goes across the sidewalk. Typically everyone's door key also opens gates and doors to community areas.

I'm not sure this helps given I don't see that GZ saw TM enter the gated community.

When Google's camera truck went by, the gates were wide open with no cars in site. The gates are left open at times and may seldom be closed.

 

[nadsat]

more papers of interest on Voice Biometrics:

Report from Sept 2011, National Biometrics Challenge from the Office of Science and Technology Policy:
http://www.biometriccoe.gov/_doc/BiometricsChallenge2011_protected.pdf

Latest draft from Investigatory Voice Biometrics Committee Report from the National Institute of Standards and Technology (NIST) dated March 9, 2012:
http://biometrics.nist.gov/cs_links/ibpc2012/presentations/Day4/IVBCReport_v1.8.pdf


2009 declassified FBI report on voice biometrics:
http://www.biometriccoe.gov/_doc/Use-Case_Committee_Report_FBI-VBS_Final_for_internet12202011.pdf

 

[Flanger]

Not sure if this is new or not. Seems the official police report has been released.

http://news.yahoo.com/blogs/lookout...rt-reveals-crime-scene-details-175656087.html

 

[curi0us]

No.

Newton: Laws of motion.
Newton: Alchemy (stupid stuff).
Therefore: Newton's laws of motion are wrong.

Dude, arguments MUST stand on their own.

Dude, do you understand what your argument actually is?

Experts: Trust us.
Randfan: Ok.
NBS: I have experience in audio engineering. I really don't think the experts can do what they are claiming in this instance. I've searched and found no science or research that supports their methods. Their actions also appear to be in violation of the standards of forensic audio examiners. One of them is using a software program ("new technology") for identification in a way the manufacturer clearly didn't intend. The other keeps saying some really woo stuff.
Randfan: Your points are invalid due to an argument from derision fallacy.

BTW: I don't even accept that it is necessarily stupid. I don't know what the context is. These are cherry picked to mock them.

If you don't know what the context is then you can't decide the quotes were cherry picked to mock them. Considering that NBS has been providing links to his sources your ignorance on context isn't something you can blame on him.

Further this "no science to back up it's claims" is nonsense.

Nobody has been able to find anything remotely close to a scientific study or double blind test that validates what these "experts" just did with their analysis of the 911 call screams. The only science available in their field refutes what they are doing.

And I want to add another "expert" to Newton. Tesla. By your logic curi0s, Tesla's claims about electricity are all wrong.

Can you really not see how obvious of a straw man that is?

When I learned physics the focus was on the science not the people behind it. Nobody ever said Newton or Tesla were "experts" so I should believe what they said.

 

[Information Analyst]

I know but I was responding to a suggestion to the possibility that the community had zero access and possibly Trayvon didn't have the code to get so maybe Zimmerman saw Trayvon hopping a fence. It would be a great excuse for Zimmerman's alleged "suspicious behavior" but I don't think that happened as every gated community I have ever lived in, had pedestrian access.

It was said previously that the vehicle entrances were open during the day, the the images on Google Street View certainly supports that. It may hang on what time at "night" they were closed, but obviously it would seem highly unlikely that Zimmerman wouldn't have mentioned it in his call to the police if he had seen Martin making an unconventional entrance!

 

[tyr_13]

What objective basis do you have to suppose that voice printing is so unreliable as to be unable to exclude GZ as it is now? Or are you speculating based on your understanding of the technology.

The confounding factors that experts previously cited as being sufficient to discount voice analysis.

To use your DNA example, it's as if the experts had said that you can't get a valid DNA sample from urine, but now a DNA sample from urine was being presented. The new expanded range of use would need testing before I was happy with it, and it would constitute attempting to falsify a revised hypothesis (or theory depending).

 

[TheL8Elvis]

Not sure if this is new or not. Seems the official police report has been released.

http://news.yahoo.com/blogs/lookout...rt-reveals-crime-scene-details-175656087.html

As far as I see, its the same partial report that has been available from the city of sanford site:
http://sanfordfl.gov/investigation/trayvon_martin.html

 

[TheL8Elvis]

Maybe it's because the only two people, as far as we can tell, on the Neighborhood Watch are violent sociopaths who are just looking for a way to boost their highly suspect virility and have no bloody clue what they are doing. If this is the same Frank Taaffe, he is a sorry sack of crap to whom no person should turn for any kind of assistsance or advice.

Really ?? Are you starting some sort of witch hunt against the neighbor now ?