Cookie consent

We use cookies to improve your experience on this website. By continuing to browse our site you agree to our use of cookies. Tell me more | Cookie preferences

Experts in Zimmerman case: Can't tell who is screaming in 911 calls

George Zimmerman (left),  Trayvon Martin

Story highlights

  • George Zimmerman trial starts Monday in Florida
  • Judge considering evidence from 911 calls
  • Audio experts testify about quality of recordings

Two days before George Zimmerman goes on trial, voice recognition experts testified it is not possible to positively identify whose screams are heard in the background of the 911 calls made the night Trayvon Martin was shot.

Florida Judge Debra Nelson will continue the hearing on the admissibility of the experts' testimony at a later date. Jury selection begins Monday.

A third audio expert was unable to testify Saturday because he was stuck on a plane.

The defense argues the state's audio experts are using unproven science to determine who is heard screaming and they shouldn't be allowed to testify. Nelson wanted to hear about the technology used for voice analysis.

Trayvon Martin shooting Fast Facts

Forensic audio expert Dr. Peter French, who testified via teleconference from the United Kingdom, described research involving the recording of stressed cries versus normal voice samples.

In the case of the 911 calls, French said that the poor quality of the recorded voices means that none of the processes available can provide any useful information about the identity of the screaming person.

"I've never come across a case in the 30 years of my career where anybody has attempted to compare screaming with normal voices," French said.

French also said in this case the recording "isn't even remotely suitable for speaker comparison."

George Doddington, an electrical engineer who works as an adviser with the National Security Agency, testified Saturday that his conclusion about the 911 call was similar to Dr. French's: that due to the quality of the data, it is not going to provide any viable source for making an identity decision.

Doddington said evaluating one second of speech to reach a conclusion is "ridiculous."

"The less you have, the worse the performance," he said.

Zimmerman, a neighborhood watch volunteer, is charged with second-degree murder in the death of Martin, 17, on February 26, 2012, in Sanford, Florida. He says he shot the teenager in self-defense.

The audio technology may be key to the prosecution's case because its experts' testimony may be able to shed light on what was said between Zimmerman and Martin moments before the teenager was shot.

Alan Reich, a voice analysis expert who testified Friday, believes that the screams recorded were not Zimmerman, but more likely came from Martin.

Ted Owens, a forensic audio engineer, testified Friday that his analysis of the 911 call indicated that the person screaming was not Zimmerman, and the person screaming is frantically asking for help in different ways before the gunshot is heard.

If the prosecution experts' analysis indicates Martin screamed for help, it could hurt the credibility of Zimmerman's claim that he acted in self-defense.

The law states that for technology to be admissible, it must be "generally accepted" in that particular field. Zimmerman's attorneys are arguing the technology does not satisfy that threshold.