IN THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF ILLINOIS EASTERN DIVISION
January 16, 2012
THADDEUS JIMENEZ, PLAINTIFF,
JEROME BOGUCKI, DEFENDANT.
The opinion of the court was delivered by: Matthew F. Kennelly, District Judge:
MEMORANDUM OPINION AND ORDER
Defendant has moved to bar the testimony of Kent Gibson. Plaintiff has offered Gibson as an expert in audio voice identification. Gibson examined an audio recording from 1993 that is said to be a recording of Ezequiel Romo speaking with Juan Carlos Torres and an audio recording of Torres speaking to authorities following his arrest around fifteen years later. He opines that it is Torres's voice on the 1993 recording.
Both sides agree that the "gold standard" for voice identification regarding a questioned recording involves comparing the recording with a known exemplar of an individual's voice, speaking at least a certain number of the words that appear on the recording. Plaintiff previously served a subpoena on Torres to appear and provide a voice exemplar. Torres is currently under indictment in state court for the murder for which plaintiff was previously convicted.
Torres objected to the voice exemplar subpoena on Fifth Amendment privilege grounds. The Court rejected this objection, as it has been established for nearly forty years that requiring a person to provide a voice exemplar does not violate his Fifth Amendment rights or, for that matter, his Fourth Amendment rights. United States v. Dionisio, 410 U.S. 1 (1973). Torres's lawyer advised, however, that if compelled to provide a voice exemplar, he would decline, meaning that he would face contempt proceedings on which he would be entitled to an immediate appeal if held in contempt.
After consulting with his expert (presumably Gibson), plaintiff's counsel advised that there was an alternate method of voice comparison that he could use that would not require a voice exemplar. Defendant's counsel said that defendant would object to use of the alternate method. The Court believed, however, that it was likely that Gibson's testimony based on the alternate method would be admissible. For this reason, and to avoid delays that might be occasioned by requiring Torres to give a voice exemplar, the Court granted Torres's motion to quash. Gibson then performed his analysis and submitted a written opinion. Defendant has moved to bar admission of the opinion at trial.
The Court has had the opportunity to review defendant's motion in limine regarding Gibson, plaintiff's response, and supporting material submitted by both sides. It now appears to the Court that it made its earlier decision improvidently, based on incomplete information. Based on the material currently before the Court, the Court concludes that plaintiff has failed to establish the reliability of Gibson's methodology and therefore grants defendant's motion in limine. Under Rule 702,
[a] witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if:
(a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles and methods; and
(d) the expert has reliably applied the principles and methods to the facts of the case.
Fed. R. Evid. 702. As the proponent of Gibson's testimony, plaintiff has the burden of establishing that his testimony is admissible. See, e.g., Lewis v. Citgo Petroleum Corp., 561 F.3d 698, 705 (7th Cir. 2009).
Gibson had to, of course, work with what was available. And what was available was not much: the 1993 recording and a recording made some fifteen years later. Gibson made his comparison based on two factors: "fundamental frequency statistical analysis," otherwise known as "average pitch analysis," and what he refers to as a "linguistic method, where accent, perceived level of education, grammar, accent, voice qualify, speaking rhythm, melody of the voice (a.k.a. prosody) and other techniques are used to compare voice." Gibson Report at 2, 3. Gibson conceded, however, that the latter is insufficient by itself and that the core basis for his opinion is the fundamental frequency statistical analysis.
Plaintiff has failed to establish that the methodology used by Gibson is sufficiently reliable to be admissible. In his written opinion, Gibson provides no support for the reliability of his methodology other than his own say-so, which is insufficient. And the deposition excerpts the parties have provided similarly offer no support for the contention that Gibson's methodology is reliable. In response to defendant's motion, plaintiff's counsel cited certain articles regarding fundamental frequency analysis, but they provided none of them to the Court, nor did they identify anything from the articles that suggests that this sort of analysis by itself (or combined with the "linguistic analysis" that Gibson also performed) is sufficient to provide a reliable voice identification. A scientific expert's analysis must be grounded in the scientific process, see Lewis, 561 F.3d at 705, and plaintiff has failed to show that is the case here.
Because plaintiff has failed to show that Gibson's analysis is sufficiently reliable to be admissible, the Court grants defendant's motion in limine and need not address the other arguments that defendant offers. The Court also does not address whether it is even possible to perform a scientific voice identification, either to include or exclude someone, by comparing the voice of a teenager, which is what Torres was in 1993, to that of an adult some fifteen years later.
MATTHEW F. KENNELLY United States District Judge
© 1992-2012 VersusLaw Inc.