Voice Biometrics Conference Convenes in NY

How topical. The annual Voice Biometrics Conference begins today in New York. This year's theme: Benchmarks, Use Cases and Real World Experience.

Registration is $699., but maybe they have media passes available. One interesting talk may be that of Alexey Khitrov, of the Speech Technology Center which is a "gold" sponsor of the conference. (Its program is SpeechPro):

Reality Check #1: Lessons Learned from Forensics and Law Enforcement

What other programs do law enforcement use for speaker recognition? [More...]

Agnitio, one of the conference's platinum sponsors, touts its Batvox and Asis programs. From reviews, they seem to be in a lead position.

Here's an interesting 2009 declassified FBI report on voice biometrics. The report notes investigatory voice biometrics may not be ready for courts --at least federal ones that use Daubert as the standard for admission of expert testimony:

The first agreement was on the meaning of the phrase - “investigatory voice biometrics”. The first word of this phrase, “investigatory”, was taken to encompass the use of speaker recognition technology in criminal and intelligence investigations and analysis.

In US Federal Courts, the admissibility of scientific evidence is determined by the presiding judge, who is guided by Federal Rules of Evidence (FRE) 104 [2] and 702 [3], among others. FRE 702 notes the Daubert Criteria, which states the following factors must be met: the technique has been tested and subjected to peer review and publication; has a known error rate and standards controlling its use; is generally accepted in the scientific community [4].

The Symposium committee members believe that automatic speaker recognition technology has not yet reached the maturity to satisfy the Daubert Criteria...Future research and evaluation are needed to advance speaker recognition technology to satisfy scientific-evidence admissibility requirements for the US Federal Courts

Also interesting:

In addition, there is a common consensus about the uncertainty of the error rates of the speaker recognition performance when operating under variations due to the speaker (e.g., stress) and variations not due to the speaker (e.g., channel distortion).

On p. 14 (its numbering) the report addresses the process for comparing a 9/11 call to a sample of a known person:

4.2 FBI Scenario 1: One-to-one Comparison

The Forensic Audio, Video, and Image Analysis Unit (FAVIAU), Operational Technology Division of the FBI, will receive a request from one of the field offices for a voice comparison examination to be conducted between an unknown voice on a 911 call and the known voice of a subject on a recording obtained by law enforcement officers.

...Currently automated comparison is performed in criminal cases solely for research purposes and only if the evidence is deemed sufficient for such an examination.

Another symposium to address the current state of voice biometrics was held in 2010. The declassified report is here, but its very technical.

Here is the latest draft (March 12, 2012) of the Investigatory Voice Biometrics Committee Report from the National Institute of Standards and Technology (NIST). On page 50-52 (its numbering) the draft addresses intelligibility, emotional state and impairment, and speaker styles (crying, chanting, etc.) and "vocal effort."

Here's the 2011 report, National Biometrics Challenge from the Office of Science and Technology Policy. It says on page 14 (image provided as it doesn't allow cut and paste):

Here is the FBI's latest on voice biometrics. As of 2008, it didn't think voice biometrics was Daubert-ready for federal courts, the standard the FBI strives to meet.

Does law enforcement use EasyVoice Biometrics? According to its website, the software was just released and didn't start shipping until March, 2012. (It costs $5,000.) I was wondering how Tom Owen (whom I have used as an expert in the past and found to be quite professional) got an advance copy since he testified at a court trial in January that he had used the program the year before. Turns out, according to Whosis, he is the administrative contact for the EasyVoice Biometrics website. If he has a proprietary or financial interest in the program, or helped develop it (which I don't know that he does or did) should he have disclosed it to the Sun Sentinel?

As for the Connecticut state court's admission of his biometric voice analysis at the prior murder trial, it should be noted that the defendant in that case was pro se, without a lawyer, and likely didn't know how to effectively challenge it. Also, it's unclear how much the analysis factored into the jury's verdict since there was DNA evidence incriminating the defendant. Her DNA was found in a blood stain on a bathroom sink handle in the victim's condo.

Jurors on Friday morning asked the court to replay testimony from Michael Bourke, a supervising DNA analyst in the state's forensic crime lab. Bourke testified the estimated frequency of someone other than Davalloo leaving that DNA profile on the sink handle was 1 in 8.5 million.

Also, the 911 case in Davaloo did not involve comparing a scream to spoken words, although I'm not sure it that matters.

On the 911 recordings, a woman tells a call-taker that a man attacked her neighbor on Harbor View Drive. The call was made from a pay phone at a fast-foot restaurant on Shippan Avenue.

The second expert hired by the media to opine on the screaming in the background of the 911 call in the Trayvon Martin-George Zimmerman shooting investigation pooh-poohed biometrics.

Ed Primeau, a Michigan-based audio engineer and forensics expert, is not a believer in the technology's use in courtroom settings. He relies instead on audio enhancement and human analysis based on forensic experience.

Hopefully whatever expert the state uses will come up with a better explanation than Primeau did for saying the voice was not Zimmerman's:

"I believe that's Trayvon Martin in the background, without a doubt," Primeau says, stressing that the tone of the voice is a giveaway. "That's a young man screaming."

On his own website, Primeau cautions that his opinion is not for legal purposes.

As an audio forensic expert I want to make clear that this is not formal voice identification and not meant for legal purposes.

He also may not have done a forensic comparison at all. He says:

If I were to conduct a formal voice identification test, here are the steps I would take:

....4. Ask for a speech exemplar from George Zimmerman screaming for help

5. Conduct a voice identification test.

So unlike Mr. Owen, Primeau thinks a test should compare a scream to a scream. Since he doesn't have an exemplar of Zimmerman screaming to compare to the 911 call with the scream in the background, he can't do a test the way he thinks it should be done.

It will be interesting to see whether the State's Attorney in Florida retains experts who use a biometric program to analyze the scream in the Trayvon Martin case, or goes with more traditional methods of speaker recognition. And whether the experts it does retain are able to draw any conclusion at all as to who screamed.

In the meantime, maybe the case will be brought up at the biometrics conference tomorrow and there will be some tweeting about it.

< The Demise of The Black Caddy | Wankers >
  • The Online Magazine with Liberal coverage of crime-related political and injustice news

  • Contribute To TalkLeft

  • Display: Sort:
    Florida is a Frye state, I think... (5.00 / 1) (#1)
    by TomMaguire on Tue Apr 03, 2012 at 05:34:31 AM EST
    IANAL, but the link suggests Florida is a Frye state.  What that would mean for a Federal trial in Florida is beyond me.

    FWIW, I checked the websites of several other expert witness firms that do audio forensics.  Everyone I found, including Tom Owen, make the same point as Mr. Primeau and stress the importance of matching speech to speech, whisper to whisper, and so on.

    I believe (none / 0) (#2)
    by jbindc on Tue Apr 03, 2012 at 06:16:52 AM EST
    Daubert superseded Frye in federal cases.

    This'll be a state case. (none / 0) (#4)
    by jpe on Tue Apr 03, 2012 at 07:16:57 AM EST
    The feds may or may not have their own trial, but the main case will be FL.  And it does indeed look like the Frye test will control.  It's been awhile since 2L Evidence, so I'm looking forward to hearing Jeralyn's take and/or the take of some of our more knowledgeable commenters.

    Yes (none / 0) (#6)
    by jbindc on Tue Apr 03, 2012 at 11:29:51 AM EST
    But Tom asked about a federal case.

    yes it did (none / 0) (#9)
    by Jeralyn on Tue Apr 03, 2012 at 12:37:39 PM EST
    yes (none / 0) (#8)
    by Jeralyn on Tue Apr 03, 2012 at 12:36:03 PM EST
    Florida uses Frye, although  bills are pending to change it.  Here's one of them. Here's more.

    My post was in two parts, the first addressing the federal Government's views on biometric voice analysis, which involves Daubert, and at the end, I brought in the Martin case, wondering if the state would seek to use biometric analysis or the more traditional voice spectrographic analysis. I wasn't really concerned with Frye or Daubert here -- a Florida appeals court approved voice spectrographic analysis in 1972 (Alea v. State, 265 So. 2d 96 (Fla. Dist. Ct. App. 3d Dist. 1972)(but noting it was being used as corroborative, not the only means of identification) but has not addressed biometric voice analysis -- so I was wondering if they'd go with a sure thing in this case or risk something new that has not been previously ruled admissible by an appeals court.


    Tom Maguire (5.00 / 2) (#3)
    by Abdul Abulbul Amir on Tue Apr 03, 2012 at 07:11:08 AM EST
    I have to admit, (none / 0) (#7)
    by RickTaylor on Tue Apr 03, 2012 at 12:09:41 PM EST
    when I first heard an expert was able to tell if the screams were Zimmerman's based on his earlier 911 calls, it sounded too good to be true. He sounded like an expert, but then when he said used something called "EasyVoice biometrics," it sounded suspicious.  I had a friend who was an applied mathematician, a genuine expert in the field of image processing, someone who's patented his own algorithms, and he did not use tools like EasyPicture Image Processing -- he rolled his own software.

    After reading Maguire's article, and unless some other authority ways in, it sounds like the hope this could all be resolved easily through scientific analysis of a recording may have been wishful thinking. We'll just have to do this the old fashioned way, weighing lots of contradictory testimony in a trial.


    Off Topic a Bit (none / 0) (#11)
    by ScottW714 on Tue Apr 03, 2012 at 02:58:45 PM EST
    But from day 1 (or day 30, whenever that call was released) mom has said it was her boy screaming.  Not sure if she could be considered an expert, or if she is telling the truth, but if she testifies it's her son's voice...

    and Zimmerman's (none / 0) (#13)
    by Jeralyn on Tue Apr 03, 2012 at 03:57:49 PM EST
    brother, father and friends will all testify it's George's voice. He also has the benefit of being overheard by the paramedic at the scene saying he had been crying out for help. And the only eyewitness to the actual struggle, John.

    Lay opinion as to voice is allowed by someone familiar with the voice.


    Other audio forensic experts (5.00 / 1) (#5)
    by TomMaguire on Tue Apr 03, 2012 at 07:27:42 AM EST
    Here are links to two other firms (one in Canada) that do audio forensics.  Finding a firm that describes its requirements at its website was not as easy as I hoped.

    Stuchman Forensic Laboratory, Advocate for Evidence Since 1992:

    Collection of Exemplar

    It is recommended that the exemplar of the known voice must be collected in as close to the same manor as the recording of the unknown voice was recorded....When the exemplar is collected, the suspect is asked by the examiner to stay the same words in the same way as they were spoken by the unknown person. In other words, in a normal, natural voice.

    Forensic Science Services of Canada


    The results depend on quality of recordings, the total number of comparison words, speakers' condition, and individual speakers' voice variability. There is a requirement for a minimum number of 20 comparison words in a `connected speech'. The suspect should provide the comparison sample by reading three times the transcript of the unknown voice sample.

    We would all like to see more experts, but so far, no one, including the two Sentinel experts,  seems to support matching screams against spoken words.

    Twin Oaks subdivision (none / 0) (#10)
    by Towanda on Tue Apr 03, 2012 at 01:01:27 PM EST
    doesn't look like it has manors.  There goes replicating results.)

    (Okay, seriously:  The links are useful, even if forensic experts lack expertise in spelling.)


    Totally Reminds Me... (none / 0) (#12)
    by ScottW714 on Tue Apr 03, 2012 at 03:07:11 PM EST
    ...of a business I went to about a decade ago that was selling secure data on their servers.  My buddy first scanned my finger, then we went downstairs, where I placed my finger on the biometric reader.  The door opens, super cool especially a decade ago.

    On the way out, my friend tells me it's all non-sense and that it's only checking for a heat signature, to make sure there is something warm on the lens.  Anyone could get it.

    Smoke and mirrors.  And now we are to buy that they can tell, without a doubt, that a voice in the background of a 911 call is definitely a person they can't get a comparison recording from.

    Seems like they, like my friends company, are trying to slap biometric on something that is anything but.

    so jeralyn, has your "subtle" hint (none / 0) (#14)
    by cpinva on Tue Apr 03, 2012 at 04:55:34 PM EST
    resulted in a free media pass yet? :)

    I.D> - Speech vs Screaming (none / 0) (#15)
    by Forensic audio expert on Fri Apr 06, 2012 at 09:58:44 AM EST
    We will have to wait until a test is done using Easy Voice Biometrics, comparing speech by a test subject, to screaming by the same subject, to see how much of a "match" there is. If the result comes out low then Mr. Owen's results were meaningless and his appearance was only a sales commercial and for self publicity.

    Time will tell. I believe the tool can be useful, but as with any lab instrument or tool, its use and results have to be well understood by the examiner, and appropriately used and interpeted.