Before Steadman, Ruiz, and Reid Associate Judges
The opinion of the court was delivered by: Ruiz, Associate Judge
Appeal from the Superior Court of the District of Columbia (Hon. Reggie B. Walton, Trial Judge)
This case, concerning the tape recording of the armed robbery of a police officer during an undercover drug operation, is before us a third time. In Taylor v. United States, 601 A.2d 1060 (D.C. 1991) (Taylor I), we held that the trial court abused its discretion by disallowing appellant, Maurice Taylor, to present a voice exemplar as demonstrative evidence to rebut that it was his voice on a police recording of the robbery without also taking the stand and submitting to cross-examination. Id. at 1066. On remand from our decision in Taylor I, the trial court ruled that the proposed comparison of the voice exemplar appellant had proffered at trial with the police tape recording was not minimally reliable and therefore inadmissible as evidence. See Taylor v. United States, 661 A.2d 636, 639 (D.C. 1995) (Taylor II). We affirmed the trial court's finding as to the unreliability of the live voice to tape comparison, but concluded that the trial court had abused its discretion by limiting its inquiry to only one exemplar format, when appellant had also proffered a tape-to-tape comparison. On remand from our decision in Taylor II, the trial court ruled that appellant's proffered exemplar of his taped voice for the purpose of comparison with the police tape - a tape-to-tape comparison - also would not be permitted because of the unreliability of the police tape for such purpose. Based on the evidentiary finding by the trial court that the police tape was unreliable, appellant filed a motion for new trial in which the police tape, which had been admitted as part of the government's case, would be excluded altogether.
Appellant appeals the trial court's denial of that motion. Because we conclude that the government did not use the police tape for the same purpose for which the defense sought to introduce it, and the defense does not otherwise challenge the admission of the police tape, we affirm.
This case arises out of an undercover police operation on July 11, 1989. Taylor was convicted of robbing Officer Dean Welch, a Metropolitan Police Department officer, at gunpoint, in the course of the officer's attempt to buy drugs from Taylor and several other individuals. See Taylor I, 601 A.2d at 1062.*fn1 During the robbery a recording device in Officer Welch's car recorded the encounter.*fn2 See id. At trial, Officer Welch recounted the robbery and identified Taylor as the gunman. The tape recording of the robbery was played for the jury. As the tape played, Officer Welch, who had prepared a transcript of the recording, identified the speakers and described their various actions. Officer Welch's in-court identification of Taylor, as well as the tape recording, figured prominently in the government's case against Taylor. See id. at 1065-66. Taylor defended on the ground that he was not present during the drug sale and robbery, and, therefore, the voice on the tape was not his. See id. at 1066. To support his defense, Taylor sought to present a live sample of his voice for the jurors to compare with the tape and decide if the voice on the tape was indeed his. See id. at 1066. The trial court rejected the proposed comparison on the grounds that Taylor's live voice exemplar would have been testimonial evidence and could be presented only if Taylor was willing to testify on the stand, under oath and subject to cross examination. See id. at 1061. Taylor chose not to testify. He was convicted of one count of armed robbery and sentenced to fifteen to forty-five years in prison. See Taylor II, 661 A.2d at 639
In Taylor I, we held that the trial court had erred by excluding the proffered live voice exemplar on the ground that it was testimonial, rather than demonstrative, evidence and remanded for the trial court to properly consider the admissibility of the proffered exemplar.*fn3 See id. at 1067. On remand, the trial court limited its inquiry to the live voice exemplar format rejected at trial and found that the voice-to-tape comparison would not meet the minimally reliable standard required for admission as evidence. See Taylor II, 661 A.2d at 639. In Taylor II, we affirmed the trial court's ruling that the voice-to-tape exemplar was not admissible but held that Taylor had intended to proffer an alternative exemplar format, a tape recording of his voice to be compared to the original police tape recording. See id. at 648. We therefore remanded the case for the trial court's consideration of the reliability of the tape-to-tape comparison with the following instructions:
If the trial court finds that this format would permit a reliable comparison with the tape of the robbery, it must grant appellant a new trial. If, however, the trial court finds that the tape-to-tape comparison would not be minimally reliable and therefore could not properly be admitted in evidence, appellant's conviction will stand affirmed, subject to his right to appeal the trial court's ruling. Id.
After our remand in Taylor II, the trial court held a hearing in which each side presented expert testimony to support their respective positions concerning the admissibility of the proposed tape-to-tape comparison. Professor Daniel Craig O'Connell, a psycholinguistics expert, testified on behalf of Taylor that the tape made at the scene of the robbery was a "low quality tape"and that a comparison would be "very difficult . . . because of the tape quality," but that the original police tape and the proffered exemplar of Taylor's voice "retain sufficient voice qualitites for a jury to make [a] judgment" if given an opportunity to hear the tapes with sufficient repetition. However, in response to the question whether a jury's level of attention and focus can adjust for the complexity of the comparison, Professor O'Connell testified, "I think it would be foolhearty (sic) to submit the tapes to the jury."
The government's expert, Bruce Koenig, a private consultant whose work involves analyzing audio and visual recordings, testified to extensive experience with voice comparison in a criminal context.*fn4 Koenig testified that the appropriate comparison process is to "look at the unknown [in this case, the police tape] to see if it is really a good enough quality to even do anything . . . then you look at the exemplar to see if it [is] good enough quality and I mean quality in the sense of recording quality, voice quality, the number of words said, how they were said. All the things you need to make a comparison." Koenig testified that the proposed exemplar tape provided by Taylor would not meet the standards promulgated by the International Association for Identification or Journal of Forensic Identification; the exemplar was not made under the same conditions as the original and possessed different noise quality and background noise, different tape speeds and frequencies. Based on his review of the proposed exemplar, Koenig concluded that one could "flip a coin and probably be more accurate than what the jury would with these particular samples." The trial court sua sponte asked Koenig: "In reference to the unknown in this case, what was the quality of that tape for comparison purposes?" During the ensuing colloquy Koenig stated that "[t]he recording quality itself [of the police tape of the robbery] even though it is certainly sufficient for intelligibility, is not very good for voice identification."
After considering the expert testimony and reviewing the tapes, the trial court concluded that the tape-to-tape format proposed by Taylor did not meet the minimally reliable standard required for admissibility of the exemplar for comparison with the police tape. See Taylor II, 661 A.2d at 648. The court reasoned that "[b]ecause the tapes are so different and therefore likely to mislead a layperson to conclude that the voice of the speakers is not the same, even if they are the same, the tape-to-tape comparison would be irrelevant on the question of who is the speaker on the tape that was recorded at the time the crime was being committed." The trial court further reasoned, agreeing with the government's expert, that "the quality of the tape that was recorded during the offense is of such poor quality that the court is compelled to agree with the government's expert's conclusion that it is inadequate for comparison purposes . . . [and] any attempt to compare it with the known tape recorded voice of the defendant would be meaningless."
Based upon the trial court's decision and "pursuant to the remand order of the Court of Appeals in [Taylor II]," Taylor filed a motion for new trial which argued that the trial court had determined that the tape of the robbery was of such poor quality that any voice identification made from the tape would lack the minimal reliability required for the admission of such evidence. Taylor claimed that the government had introduced just such a voice identification by Officer Welch at trial. According to the motion, a new trial should be granted because "[i]f evidence is unreliable for the defense, surely it is equally unreliable for the government." [Id.] The court denied Taylor's motion for new trial, deeming it a collateral attack on his conviction and holding that Taylor failed to meet the "cause and prejudice" standard set forth in Head v. United States, 489 A.2d 450, 451 (D.C. 1985). Additionally, reaching the merits despite its holding pursuant to Head, the court noted that the government had not used the tape recording as an identification procedure, but only as proof that the robbery occurred, and that "the defendant apparently misconstrues the purpose for which the evidence was admissible." According to the trial court:
Although the robbery victim said that the voice on the tape recording was the defendant's voice, that can hardly be said to amount to an identification of the defendant. And although the Court of Appeals referenced what the victim said about the voice on the audio recording when it discussed the strength of the government's identification evidence, the essence of the Court of Appeals' conclusion about the strength of the victim's identification related to his visual observations, his description of the robber and his subsequent visual identifications of the defendant.
. . . There was no comparison for identification purposes conducted in the jury's presence, or testimony that a comparison was conducted outside of the jury's presence. Moreover, the defendant did not testify during the trial; accordingly, the jury never heard the defendant speak. Therefore, there was no basis upon which the jury could have factored the victim's reference to the voice on the audio tape into its analysis of whether the government proved that the defendant was the robber.
The trial court then held that the tape was properly admitted as relevant evidence which provided "corroborative proof ...