Trayvon Martin, Vigilante Justice

[Unabogie]

Here is a link to a youtube video giving a demonstration on how Easy Voice works.

pt 1


pt 2


pt 3


ETA: apparently I can't count because it's three videos.

Thank you. That last video seems like a pretty good refutation of the hand waving going on.

 

[Moss]

^^^ This 100X ^^^! One of the things that makes me instantly skeptical here is that real scientists are extremely unlikely to render an opinion without enough information and it's almost always those that are not practicing good science that are eager to rush in and make pronouncements far too early.

I am willing to be wrong, but I find it highly doubtful that anyone can come to meaningful conclusions based on the recordings we've heard.

That might depend on the model and the algorithms they use. I find it amusing how fast "woo!" and "questionable" is bandied about. And how a twenty five year old paper on a different model of voice identification is used as prime evidence against a different technique.
I'm not qualified to speak whether it is possible to do what is claimed for that software. How qualified are you?
The arguments I've heard so far make me think of argument from incredulity. Whether that is a warranted position to take depends not on what you believe is possible but what the software actually shows during impartial tests. I'd rather wait and see what the court and other experts make of this than proclaim that software woo on an insufficient basis.

 

[crimresearch]

Thanks! No ****! Mind boggling that people are arguing otherwise.

Other than being completely wrong, that's almost right.

 

[Skeptic Ginger]

I'm with WildCat. Having done spectrometric pattern matching a few times over the years, it seems unlikely that the current state of technology can arrive at an accurate assessment under these conditions.

Speech recognition relies on acoustics unique to a person's physiology. To believe that the circumstances of the recording would not distort those signatures outside of all recognition is far-fetched.

Let's see how Owen's qualifications compare to "having done it a few times over the years":

From the paper that hired him:

A leading expert in the field of forensic voice identification sought to answer that question by analyzing the recordings for the Orlando Sentinel.
Tom Owen, forensic consultant for Owen Forensic Services LLC and chair emeritus for the American Board of Recorded Evidence, used voice identification software to rule out Zimmerman. Another expert contacted by the Sentinel, utilizing different techniques, came to the same conclusion.

American Board of Recorded Evidence

The American College of Forensic Examiners International is an independent, scientific, and professional association representing forensic examiners worldwide.

Multi-disciplinary in its scope, the college actively promotes the dissemination of forensics information and the continued advancement of forensic examination and consultation across the many professional fields of membership. The college has elevated standards through education, credentials, basic and advanced training, as well as Diplomate and Fellow status.

I looked to see if this wasn't just a front group established by Owen himself to give more legitimacy to his company. It is not. It was founded by Robert O'Block, PhD, PsyD; American College of Forensic Examiners International

Advisory Boards

We oversee 12 advisory boards focusing on individual areas within forensic science. ACFEI currently has twelve advisory boards that relate to the many aspects of forensics. Each board has unique requirements and strict standards. ACFEI members have the opportunity to become a Diplomate of any of these boards provided they have the experience and knowledge to do so.

Please select a board below to learn more about the standards and requirements and to meet the current board members (displayed in new window).

» American Board of Certification in Infrastructure Protection (formerly American Board for Critical Incident Professionals)—Moved to ABCHS.
» American Board of Forensic Accounting
» American Board of Forensic Counselors
» American Board of Forensic Dentistry
» American Board of Forensic Examiners
» American Board of Forensic Engineering and Technology
» American Board of Forensic Medicine
» American Board of Forensic Nursing
» American Board of Forensic Social Workers
» American Board of Psychological Specialties
» American Board of Law Enforcement Experts—Moved to ABCHS.
» American Board of Recorded Evidence
» American Board of Registered Investigators

Credentialing Programs notes 11 areas one can be certified in. This is very similar to nursing specialty certifications and medical board certifications.

Accreditation

ACFEI is an approved provider of Continuing Education by the following:

» Accreditation Council for Continuing Medical Education (ACCME)
» National Association of State Boards of Accountancy (NASBA)
» National Board for Certified Counselors (NBCC)
» California Board of Registered Nursing (CBRN)
» American Psychological Association (APA)
» California Board of Behavioral Sciences (CBBS)
» Association of Social Work Boards (ASWB)
» American Dental Association (ADA CERP)

They publish a peer reviewed journal of forensic science


The bottom line here, Owen is the Chair Emeritus: Thomas J. Owen, BA, FACFEI, DABRE, CHS-V (retired chair).


I already mentioned Owen's publications. His work has been published in the Audio Engineering Society:

The Journal of the Audio Engineering Society — the official publication of the AES — is the only peer-reviewed journal devoted exclusively to audio technology. Published 10 times each year, it is available to all AES members and subscribers.

The Journal contains state-of-the-art technical papers and engineering reports; feature articles covering timely topics; pre and post reports of AES conventions and other society activities; news from AES sections around the world; Standards and Education Committee work; membership news, patents, new products, and newsworthy developments in the field of audio.

I think we can see from this that the person hired by the Orlando Sentinel was not some forensics analyst out of the yellow pages. This guy really is at the top of this specific field.


Oh, but you and WC dabbled in this stuff, so you know better.

 

[Skeptic Ginger]

Well, if the guy with software to sell decades of experience, multiple national certifications in audio forensic science, a frequent speaker at national and international forensics conferences, dozens of published papers on audio forensics, and who is recognized as one of the top leaders in audio recognition forensics says so.

ftfy

 

[WildCat]

Then why haven't you picked up on the fact that the study was for spectrographic voice identification and the scream was analyzed using biometric voice identification?

It's the same damned thing!

 

[Unabogie]

Let's see how Owen's qualifications compare to "having done it a few times over the years":

From the paper that hired him:
American Board of Recorded EvidenceI looked to see if this wasn't just a front group established by Owen himself to give more legitimacy to his company. It is not. It was founded by Robert O'Block, PhD, PsyD; American College of Forensic Examiners International

Advisory Boards

Credentialing Programs notes 11 areas one can be certified in. This is very similar to nursing specialty certifications and medical board certifications.

Accreditation

They publish a peer reviewed journal of forensic science


The bottom line here, Owen is the Chair Emeritus: Thomas J. Owen, BA, FACFEI, DABRE, CHS-V (retired chair).


I already mentioned Owen's publications. His work has been published in the Audio Engineering Society:

I think we can see from this that the person hired by the Orlando Sentinel was not some forensics analyst out of the yellow pages. This guy really is at the top of this specific field.


Oh, but you and WC dabbled in this stuff, so you know better.

Sgtbaker provided video of how the software works. I'm not sure what they are disputing about the technique, since the video covers poor recording, altered voices, and recordings with noise or music.

People should also just download "sound hound", to see how far sound parsing has come in recent years.

 

[Skeptic Ginger]

What do you mean by having "access to a clean copy of the recording"? The Orlando Sentinel article said nothing about this. Are you saying that the Sanford Police gave the Orlando Sentinel their own "clean" (whatever that means) copy of the recording?

The news media get a copy directly from the 911 dispatchers. Typically the police make a bunch of cds and hand them out to legitimate news sources. The rest of us all hear those copies played on TV, radio and Youtube.

 

[Natural Born Skeptic]

That might depend on the model and the algorithms they use. I find it amusing how fast "woo!" and "questionable" is bandied about. And how a twenty five year old paper on a different model of voice identification is used as prime evidence against a different technique.
I'm not qualified to speak whether it is possible to do what is claimed for that software. How qualified are you?
The arguments I've heard so far make me think of argument from incredulity. Whether that is a warranted position to take depends not on what you believe is possible but what the software actually shows during impartial tests. I'd rather wait and see what the court and other experts make of this than proclaim that software woo on an insufficient basis.

Well, I'm quite openly saying:
1. I'm skeptical.
2. I don't know and could be wrong (about my skepticism). But detection of woo is very strong in me and if I had to wager money I would wager it on being right .
The problem/challenge is that the term "expert" means nothing to me and I don't care what courts decide either. I want to know what expert scientists say about this. And the problem with a lot of these fields is that they are so bogus that real scientists don't even pay much attention to them. So it's a real challenge to get good information on it (I've Googled quite a bit on it). And in fact, that is one of the things that makes me quite skeptical as well. If this was a well established science it seems like we should be able to find a ton of peer reviewed research papers on it.

 

[WildCat]

Okay, Mister True Skeptic™. So where does uncritically accepting Zimmerman's uncorroborated self-defense claim that Martin attacked him while he was walking back to his truck fall on the scale of True Skepticism™?

Who did that? We don't have the data to know either way.

 

[Checkmite]

Where had WC done that?

I admit it's merely a suspicion; it's just a bit odd that he (and in fairness not just he) comes up with a special reason why every single thing that casts doubt upon Zimmerman's story doesn't or shouldn't count; but things that don't cast any such doubt, don't get the same scrutiny. A witness who sees Zimmerman being pummeled is reliable; one who doesn't see that is questionable. One can tell by ear that the crying voice is Zimmerman based on what we know his voice sounds like from the 911 call; but when one tries to use technology to accomplish the same end it's "woo". Trayvon's girlfriend has been proven to have been on the phone exactly when she claimed to be; but the remainder of her testimony is "not credible" anyway. Zimmerman thinking Martin "suspicious" because he was black like the suspect in some burglary months prior and choosing to follow him by vehicle and on foot while describing him using expletives on the phone with police was reasonable; Martin reacting negatively in any way to being followed by some stranger in the middle of the night who continued to pursue even after Martin ran is irrational and makes him the aggressor. And so on, and so forth.

 

[WildCat]

Well, I'm quite openly saying:
1. I'm skeptical.
2. I don't know and could be wrong (about my skepticism). But detection of woo is very strong in me and if I had to wager money I would wager it on being right .
The problem/challenge is that the term "expert" means nothing to me and I don't care what courts decide either. I want to know what expert scientists say about this. And the problem with a lot of these fields is that they are so bogus that real scientists don't even pay much attention to them. So it's a real challenge to get good information on it (I've Googled quite a bit on it). And in fact, that is one of the things that makes me quite skeptical as well. If this was a well established science it seems like we should be able to find a ton of peer reviewed research papers on it.

But but but the marketing department renamed it biometric analysis! Therefore it magically becomes as accurate as DNA or eye retina identification!!1!!!!!11!eleventy!1!111!1!!

 

[Natural Born Skeptic]

Here's what I'd like to see. Rigorously conducted double blind blind tests showing the use of voice identification software successfully identifying matches, simulating a situation like the one here where we've got voices in the background on a 911 tape. Is anyone aware of such tests??

 

[Moss]

Well, I'm quite openly saying:
1. I'm skeptical.
2. I don't know and could be wrong (about my skepticism). But detection of woo is very strong in me and if I had to wager money I would wager it on being right .
The problem/challenge is that the term "expert" means nothing to me and I don't care what courts decide either. I want to know what expert scientists say about this. And the problem with a lot of these fields is that they are so bogus that real scientists don't even pay much attention to them. So it's a real challenge to get good information on it (I've Googled quite a bit on it). And in fact, that is one of the things that makes me quite skeptical as well. If this was a well established science it seems like we should be able to find a ton of peer reviewed research papers on it.

Ok, that I can agree on because I had the same trouble finding relevant papers pertaining to that. But then again the problem might be that I used incorrect terms for that specific field.
Part of the problem with that I suspect that is the result of that software being a commercial product. So are most other voiceprinting applications. So I'd rather look at patent applications in the field than at papers.

 

[Dancing David]

http://articles.orlandosentinel.com...easonable-scientific-certainty?pagewanted=all

And without the data analysis and database to demonstrate that, it is as of yet an unsupported assertion.

 

[Cylinder]

I admit it's merely a suspicion;

If you were a juror and know what you know at this point from this discussion, would you convict or aquit?

 

[Natural Born Skeptic]

Moss, do you agree per my post above that double blind tests with the tester being able to correctly make matches are an absolute must if we are to give credibility to this field? How can we trust it with anything less?

 

[WildCat]

I admit it's merely a suspicion; it's just a bit odd that he (and in fairness not just he) comes up with a special reason why every single thing that casts doubt upon Zimmerman's story doesn't or shouldn't count; but things that don't cast any such doubt, don't get the same scrutiny.

Where does the burden of proof lie Checkmite?

A witness who sees Zimmerman being pummeled is reliable; one who doesn't see that is questionable.

Well yes, an eyewitness is more reliable than a non-eyewitness. The eyewitness heard and saw the scuffle, as opposed to just hearing it.

One can tell by ear that the crying voice is Zimmerman based on what we know his voice sounds like from the 911 call; but when one tries to use technology to accomplish the same end it's "woo".

No, the eyewitness actually saw the person screaming, and since we have 2 ears it's quite easy to tell where sound is coming from, which is the very reason evolution provided us with 2 ears. And the eyewitness can actually see lips move as well as the sound and doesn't have to make assumptions like Cutcher did.

Trayvon's girlfriend has been proven to have been on the phone exactly when she claimed to be; but the remainder of her testimony is "not credible" anyway.

When did I say this? I never said her testimony was not credible, except when she said things she couldn't possibly know from hearing it on the phone, such as who pushed who first. If she could do that Randi has $1 million waiting for her.

Zimmerman thinking Martin "suspicious" because he was black like the suspect in some burglary months prior and choosing to follow him by vehicle and on foot while describing him using expletives on the phone with police was reasonable; Martin reacting negatively in any way to being followed by some stranger in the middle of the night who continued to pursue even after Martin ran is irrational and makes him the aggressor. And so on, and so forth.

Neither party to this had the right to lay hands on the other, no matter what their suspicions were. Once that happens the other party is engaging in self-defense.

 

[Dan O.]

It's the same damned thing!

It's two different approaches to the same goal.

 

[WildCat]

It's two different approaches to the same goal.

How do the approaches differ Dan?