dissenting.
For the reasons stated in the majority opinion, I agree that the voice identification procedure was unduly suggestive; however, I disagree with the majority’s finding that the identification nonetheless was reliable.
“Suggestive confrontations are disapproved because they increase the likelihood of misidentification.” Neil v. Biggers, 409 U.S. 188, 198, 93 S.Ct. 375, 382, 34 L.Ed.2d 401 (1972).
The constitutional safeguards that apply to visual identifications also apply to voice identifications. [Thus, a] “voice identification is inadmissible if its reliability is outweighed by the suggestiveness of the identification procedure.” “Reliability depends on such factors as the witness’s opportunity to hear the accused and the consistency with prior voice identifications.” In other words, the witness must have an adequate basis for comparing defendant’s voice with the voice he or she identified as the voice of the assailant.
State v. Gallagher, 286 N.J.Super. 1, 668 A.2d 55, 62-63 (App.Div.1995) (citations omitted); see also United States v. Schultz, 698 F.2d 365 (8th Cir.1983). “It is the likelihood of misidentification which violates a defendant’s right to due process.” Biggers, 409 U.S. at 198, 93 S.Ct. at 381-82. Thus, a voice identification is inadmissible when the evidence establishes a substantial likelihood of misidentification. See id.
The record establishes a substantial likelihood of misidentification by Leslie Lavell. The record fails to prove that Lavell had the opportunity to give a significant degree of attention to the voice she heard on the telephone. See id. at 199, 93 S.Ct. at 382 (stating that one of the factors to be considered in evaluating the likelihood of misidentification is the “witness’ degree of attention”). She received only one telephone call *477from the person who made the threat and did not specifically indicate the length of the call. Lavell’s description of the words spoken suggests, however, that the call was very brief. She testified that “the person on the phone said, it’s a bomb in the MF building and then hung up.” Although Lavell reported to the police that the caller was an African-American male, she did not otherwise describe the caller. She did not say whether the voice was low or high pitched, deliberate, excited, or otherwise distinctive.
In addition to the brevity of the call, the other evidence also detracts from Lavell’s claim of certainty of identification. See id. (stating that another factor is “the level of certainty demonstrated by the witness at the confrontation”). Within a half-hour after she received the call, a police officer asked Lavell to listen to a recording of a person’s voice. She did, but could not recognize the voice. The evidence proved Dance’s voice was on that recording. Thus, Lavell could not identify Dance’s voice shortly after the call and within the same context of the call. Her failure to do so undermines both her initial degree of attention and the level of certainty that she later claimed to have regarding the voice identification.
When the officer entered Tarra Hendren’s residence, three people were then present — Dance, a woman, and a man who stutters. While making an audio recording of their conversations, the officer confronted the three people and accused Dance of making the telephone call. When the officer took that recording to Lavell, she understood from the officer’s comments that the recording concerned the bomb threat she had received. She also knew the recording had been made after she received the telephone call. As Lavell listened to the recording, she heard the officer make the accusation that someone from the house had called in a bomb threat to her. Clearly, the likelihood of “misidentification is ... heightened if the police indicate to the witness that they have other evidence that one of the persons ... committed the crime.” Simmons v. United States, 390 U.S. 377, 383, 88 S.Ct. 967, 971, 19 L.Ed.2d 1247 (1968). Moreover, no evidence tends to *478prove that the time that elapsed between the telephone call to Lavell and the officer’s arrival at Hendren’s residence was so short that it was unlikely another person had left the residence before the officer arrived.
When the officer began “conversing” with Dance, the audio recording captured Dance’s protest. Lavell testified that, when she heard a man “raising his voice ... [, saying] they were harassing him when they were trying to say that he did it,” she identified that person as the caller. Lavell did not testify, however, that the caller who made the bomb threat raised his voice.
“The suggestive elements in this identification procedure made it all but inevitable that [Lavell] would identify [Dance] whether or not he was in fact “the man.” In effect, the police repeatedly said to [Lavell], “This is the man.” This procedure so undermined the reliability of the ... identification as to violate due process.”
Foster v. California, 394 U.S. 440, 443, 89 S.Ct. 1127, 1129, 22 L.Ed.2d 402 (1969).
The totality of these circumstances does not support a finding of reliability. The evidence fails to prove that Lavell had an adequate basis for making the comparison between the caller and Dance’s voice. One half-hour after the call, she could not identify Dance’s voice as the caller. Later, the police essentially told Lavell the second recording would have the voice of the caller. The recording on which Lavell claims to have identified the caller contained voices of Dance, a woman and a man who stutters. The voices of the woman and the man who stutters would have been distinctive and were not consistent with the voice of the caller, as initially described by Lavell. In other words, it was “all but inevitable,” id., that Lavell would ehminate the voices of the woman and the stuttering man and identify the remaining voice as the caller. Indeed, Lavell testified that she could eliminate the voice of the stuttering man because she knew him.
In addition, the conclusion is unavoidable that, when Lavell heard Dance’s voice on the first recording and could not *479identify that voice as the caller, the repetition of his voice on the second recording, after the suggestive representations, made it inevitable that she would then say Dance was the caller. The evidence leaves to speculation whether Lavell had the ability to discern that the second recording contained a voice familiar to her because it was indeed the caller or, rather, merely because it was the voice she heard on the first recording. The substantial likelihood of misidentification flowing from this circumstance alone is compounded by the officer’s representation to her that the second recording was made at the house where the call originated and that one of the voices was that of the caller.
The explanation for the identification is inextricably linked to the suggestive representations the police made to Lavell. Those suggestive representations undermine the credibility of the identification and establish more than a substantial likelihood of misidentification. For these reasons, I would hold that the trial judge erred in refusing to suppress the voice identification. Accordingly, I would reverse the conviction and remand for a new trial.