Reed v. State

391 A.2d 364, 283 Md. 374, 97 A.L.R. 3d 201, 1978 Md. LEXIS 426

• Court: Court of Appeals of Maryland
• Decided: September 6, 1978
• Docket: [No. 62, September Term, 1977.]
• Status: Published

Opinions

Eldridge, J.,

delivered the opinion of the Court. Murphy, C. J., and Smith and Orth, JJ., dissent and Smith, J., filed a dissenting opinion in which Murphy, C. J., and Orth, J., concur at page 400 infra.

The issue in this criminal case is the admissibility of voice identification testimony based on the analysis of spectrograms, commonly described as “voiceprints.”

In September 1974, a woman was raped, late at night, outside her home in Montgomery County, Maryland. She immediately entered a hospital for treatment and reported the incident to the police. The following afternoon, she received a telephone call from a person who identified himself as her assailant. The victim notified the Montgomery County Police Department, and the police attached a recording device to her telephone. During the next three days, the victim received and recorded seven telephone calls, all apparently placed by the original caller.

During the course of one of these telephone conversations, the victim’s caller asked to have intercourse with her again. She offered instead to pay him $1,000.00. In a subsequent conversation, she and the caller arranged for her to deliver $1,000.00 to the locker room of the Greyhound Bus Station in the District of Columbia. She was to find the key of locker number 326 on top of an electrical “plug” box, place the money inside the locker, and return the key to its original location on the box. The victim then complied with her caller’s instructions. Afterwards, the defendant James Reed appeared at the bus station, entered the locker room, picked up the key from the box, and proceeded toward locker 326. As he approached the locker, police officers, who had been watching the locker room from a hole drilled in the door between the locker and boiler rooms, emerged from the boiler room and arrested Reed. Reed was subsequently indicted on rape and other charges growing out of the same incident.

In May 1975, Reed was compelled to submit voice exemplars to the State’s Attorney. Reed was required to repeat, into a telephone connected to a recording device, the words spoken to the victim by her assailant in the September 1974 telephone calls. These tapes, together with a composite recording of the calls made by the assailant, were then sent to the Voice Identification Unit of the Michigan State Police Department for speetrographic analysis and comparison. The results of this comparison were considered inconclusive, and in August 1975 Reed was required to submit another set of voice exemplars, again reading the words spoken by the assailant. These voice samples were also sent to Michigan for speetrographic analysis and comparison. This second test resulted in an alleged positive identification of Reed as the speaker on four of the seven calls made by the rapist.¹

A pretrial suppression hearing on the admissibility into evidence of voice identification testimony based on spectrographic analysis was conducted in the Circuit Court for Montgomery County. After hearing evidence on the general validity and reliability of the spectrographic method of identification, the trial court ruled that the State could at Reed’s criminal trial introduce expert testimony based on spectrographic analysis for the purpose of voice identification.

Reed was first brought to trial in October 1975. Voiceprint testimony was introduced, purportedly identifying Reed as the speaker who had placed the calls to the victim. After two and one-half days of deliberation, the jury was unable to reach agreement, and a mistrial was declared. In March 1976, Reed was again brought to trial, and voiceprint testimony was again introduced. In this second trial, Reed was found guilty of rape, unnatural and perverted sex acts, robbery, verbal threats, and unlawful use of the telephone. Reed was sentenced to life imprisonment for the crime of rape and to lesser concurrent terms of imprisonment for the remaining crimes. The Court of Special Appeals affirmed the convictions, Reed v. State, 35 Md. App. 472, 372 A. 2d 243 (1977). This Court then granted Reed’s petition for a writ of certiorari to consider the trial court’s admission of voiceprint evidence. We hold that the admission of this evidence was error.²

The voiceprint technique, although of relatively recent origin, has been much discussed and described in cases and legal commentaries.³ The process involves the use of a machine known as a spectrograph. This machine analyzes the acoustic energy of the human voice into three components — time, frequency, and intensity — and graphically displays these components by generating, through an electric stylus, a series of closely spaced light and dark lines, varying in position, on a sheet of electrically sensitive paper. The resulting graphic representation is what is called a spectrogram or “voiceprint.” It reveals certain patterns or “formats” which correspond to the sounds which are analyzed. According to the testimony of Dr. Oscar Tosi, the State’s principal witness at the suppression hearing and the most widely known proponent of the reliability of the voiceprint technique, ⁴ spectrography

“consists of comparing both aurally and visually spectrograms of a questioned voice and a known voice, and on the basis of the similarities to decide whether or not the two voices, the questioned and known voice are the same or belong to different persons.”

Essentially, therefore, the task of spectrography is one of pattern matching. It is dependent on the individual judgment of the examiner. As stated by Dr. Tosi:

“I consider [spectrography] reliable only if the examiner is reliable and he adjusts to what the conditions are, and he is an honest person; and he is prone to use no opinion, but probability rather than positive identification in cases of some doubt. Then I consider this good. It is reliable and can be used only ... under these circumstances. Otherwise it would be a disaster.”

The examiner’s task is complicated by what is termed “intra-speaker” variability, that is, the fact that individual speakers apparently do not say the same word in precisely the same way each time they utter it, and that spectrograms reflect this difference. According to Detective Sergeant Lonnie Smrkovski of the Michigan State Police, the examiner in the instant case, if a speaker were to utter the same word on fifty consecutive days, he would expect none of the resulting spectrograms to be identical.⁵ It is maintained, however, that the differences between the separate utterances of an individual speaker are less than the differences between the utterances of different speakers, so that intra-speaker variations do not render identification impossible.

According to Sgt. Smrkovski, at least ten points of similarity must be noted between two speech samples before a positive identification can be achieved. Apparently, this is independent of the number of the speech samples being compared. In the instant case, Sgt. Smrkovski, after listening to the tapes submitted to him, selected 138 of the 2,162 words spoken for comparison and made spectrograms of these words. In this sample, Sgt. Smrkovski rated one comparison “excellent,” twenty “very good,” thirty-seven “good,” and thirty-five “fair.” These comparisons were the basis of his conclusion that Reed’s voice and the voice of the victim’s caller were the same.

A principal consideration with regard to the admissibility of expert testimony, according to Wigmore, is: “On this subject can a jury receive from this person appreciable help?” 7 Wigmore, Evidence § 1923 (Chadbourn rev. 1978). Clearly, this is dependent on the particular circumstances of each case. No rule or set of rules could be expressed for all cases which would adequately distinguish helpful expert testimony from that which is superfluous or worse. Accordingly, this Court has held that the determination of similar and related issues are generally matters within the sound discretion of the trial court. Beahm v. Shortall, 279 Md. 321, 340, 368 A. 2d 1005 (1977); Greenstein v. Meister, 279 Md. 275, 283, 368 A. 2d 451 (1977); Radman v. Harold, 279 Md. 167, 168, 367 A. 2d 472 (1977), and cases there cited.

On the other hand, with particular regard to expert testimony based on the application of new scientific techniques, it is recognized that prior to the admission of such testimony, it must be established that the particular scientific method is itself reliable. People v. Kelly, 17 Cal. 3d 24, 130 Cal. Rptr. 144, 549 P. 2d 1240 (1976); Jones, Danger — Voiceprints Ahead, 11 Am. Crim. L. Rev. 549, 554 (1973). See also Shanks v. State, 185 Md. 437, 440, 45 A. 2d 85 (1945); 3 Wigmore, Evidence § 795 (Chadbourn rev. 1970).

On occasion, the validity and reliability of a scientific technique may be so broadly and generally accepted in the scientific community that a trial court may take judicial notice of its reliability. Such is commonly the case today with regard to ballistics tests, fingerprint identification, blood tests, and the like. See Shanks v. State, supra, 185 Md. at 440. Similarly, a trial court might take judicial notice of the invalidity or unreliability of procedures widely recognized in the scientific community as bogus or experimental. However, if the reliability of a particular technique cannot be judicially noticed, it is necessary that the reliability be demonstrated before testimony based on the technique can be introduced into evidence. Although this demonstration will normally include testimony by witnesses, a court can and should also take notice of law journal articles, articles from reliable sources that appear in scientific journals, and other publications which bear on the degree of acceptance by recognized experts that a particular process has achieved. People v. Law, 40 Cal.App.3d 69, 75, 114 Cal. Rptr. 708, 711 (1974).

The question of the reliability of a scientific technique or process is unlike the question, for example, of the helpfulness of particular expert testimony to the trier of facts in a specific case. The answer to the question about the reliability of a scientific technique or process does not vary according to the circumstances of each case. It is therefore inappropriate to view this threshold question of reliability as a matter within each trial judge’s individual discretion. Instead, considerations of uniformity and consistency of decision-making require that a legal standard or test be articulated by which the reliability of a process may be established.

The test which has gained general acceptance throughout the United States for establishing the reliability of such scientific methods was first articulated in the leading case of Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923):

“Just when a scientific principle or discovery crosses the line between the experimental and demonstrable stages is difficult to define. Somewhere in this twilight zone the evidential force of the principle must be recognized, and while courts will go a long way in admitting expert testimony deduced from a well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs." (Emphasis supplied.)

That is to say, before a scientific opinion will be received as evidence at trial, the basis of that opinion must be shown to be generally accepted as reliable within the expert’s particular scientific field. Thus, according to the Frye standard, if a new scientific technique’s validity is in controversy in the relevant scientific community, or if it is generally regarded as an experimental technique, then expert testimony based upon its validity cannot be admitted into evidence.

The identity of the relevant scientific community is, of course, a matter which depends upon the particular technique in question. In general, members of the relevant scientific community will include those whose scientific background and training are sufficient to allow them to comprehend and understand the process and form a judgment about it. In unusual circumstances, a few courts have held that the experts thus qualified might properly be from a somewhat narrower field. See People v. Williams, 164 Cal.App.2d Supp. 858, 331 P. 2d 251 (1958).

This criterion of “general acceptance” in the scientific community has come to be the standard in almost all of the courts in the country which have considered the question of the admissibility of scientific evidence. See, e.g., Rivers v. Black, 259 Ala. 528, 68 So. 2d 2 (1953); Pulakis v. State, 476 P. 2d 474 (Alas. 1970); State v. Valdez, 91 Ariz. 274, 371 P. 2d 894 (1962); People v. Busch, 56 Cal. 2d 868, 16 Cal. Rptr. 898, 366 P. 2d 314 (1961); People v. Williams, supra; Brooke v. People, 139 Colo. 388, 339 P. 2d 993 (1959); Kaminski v. State, 63 So. 2d 339 (Fla. 1953); Salisbury v. State, 221 Ga. 718, 146 S.E.2d 776 (1966); State v. Linn, 93 Idaho 430, 462 P. 2d 729 (1969); State v. Lowry, 163 Kan. 622, 185 P. 2d 147 (1947); State v. Casale, 150 Me. 310, 110 A. 2d 588 (1954); Commonwealth v. Fatalo, 346 Mass. 266, 191 N.E.2d 479 (1963); People v. Morse, 325 Mich. 270, 38 N.W.2d 322 (1949); State v. Kolander, 236 Minn. 209, 52 N.W.2d 458 (1952); State v. Stout, 478 S.W.2d 368 (Mo. 1972); Boeche v. State, 151 Neb. 368, 37 N.W.2d 593 (1949); State v. Arnwine, 67 N.J. Super. 483, 171 A. 2d 124 (1961); State v. Trimble, 68 N. M. 406, 362 P. 2d 788 (1961); People v. Alston, 79 Misc. 2d 1077, 362 N.Y.S.2d 356 (1974); State v. Steele, 27 N.C. App. 496, 219 S.E.2d 540 (1975); State v. Swanson, 225 N.W.2d 283 (N.D. 1974); State v. Smith, 50 Ohio App.2d 183, 362 N.E.2d 1239 (1976); Henderson v. State, 94 Okla. Crim. 45, 230 P. 2d 495 (1951); State v. Green, 271 Ore. 153, 531 P. 2d 245 (1975); United States v. Bruno, 333 F. Supp. 570 (E.D. Pa. 1971); Romero v. State, 493 S.W.2d 206 (Tex. Crim. App. 1973); State v. Woo, 84 Wash. 2d 472, 527 P. 2d 271 (1974); Puhl v. Milwaukee Automobile Ins. Co., 8 Wis. 2d 343, 99 N.W.2d 163 (1959).⁶

Although Frye v. United States, supra, was a case involving the results of a lie detector examination, the test itself has been broadly applied, and judged the appropriate standard to apply to newly developed methods of scientific discovery. The Frye test has been invoked by courts in their consideration of, inter alia, paraffin test, Brooke v. People, supra; medical testimony regarding the cause of birth defects, Puhl v. Milwaukee Automobile Ins. Co., supra; breath analysis devices designed to test for intoxication, People v. Morse, supra; truth serum injections, State v. Linn, supra; blood tests, People v. Alston, supra; neutron activation analysis, State v. Stout, supra; gunshot residue tests, State v. Smith, supra; Nalline tests for detection of narcotics use, People v. Williams, supra; ink identification tests, United States v. Bruno, supra; and hypnotism, People v. Busch, supra.

This Court in Shanks v. State, supra, although not citing the Frye case itself, recognized the standard of general scientific acceptance in connection with the admissibility of blood test evidence. Chief Judge Marbury there pointed out (185 Md. at 440, emphasis supplied):

“In the early cases evidence of the tests was not admitted, because the courts here were not convinced of their general acceptance and reliability. See State v. Damm, 62 S. D. 123, 252 N. W. 7; Beuschel v. Manowitz, 241 App. Div. 888, 272 N. Y. S. 165. Blood tests are now accepted everywhere, scientifically, as accurate, and the courts . . . have generally followed the same view.”

Almost every state court that has considered voiceprint evidence in a reported opinion has applied the Frye or a similar standard in determining the question of its admissibility. See Hodo v. Superior Court, 30 Cal.App.3d 778, 784, 106 Cal. Rptr. 547, 550 (1973); People v. Kelly, 17 Cal. 3d 24, 130 Cal. Rptr. 144, 549 P. 2d 1240 (1976); People v. Law, supra; People v. King, 266 Cal.App.2d 437, 72 Cal. Rptr. 478 (1968); Brown v. United States, 384 A. 2d 647 (D.C. C.A. 1978); Worley v. State, 263 So. 2d 613, 614 (Fla. App. 1972); Commonwealth v. Lykus, 367 Mass. 191, 327 N.E.2d 671, 678 (1975); People v. Tobey, 401 Mich. 141, 257 N.W.2d 537 (1977); State v. Cary, 99 N.J. Super. 323, 239 A. 2d 680, 685 (1968), aff’d, 56 N. J. 16, 264 A. 2d 209 (1970); D’Arc v. D’Arc, 157 N.J. Super. 553, 385 A. 2d 278 (1978); People v. Rogers, 86 Misc. 2d 868, 385 N.Y.S.2d 228, 237 (1976); State v. Olderman, 44 Ohio App.2d 130, 336 N.E.2d 442, 448 (1975); Commonwealth v. Topa, 471 Pa. 223, 369 A. 2d 1277, 1281 (1977). Contra, State ex rel. Trimble v. Hedman, 291 Minn. 442, 192 N.W.2d 432 (1971) (scientific disagreement goes to weight, not admissibility); see also Alea v. State, 265 So. 2d 96 (Dist. Ct. App. Fla. 1972) (issue not discussed).

The Frye test has been subjected to some criticism, primarily on the grounds that it is too conservative and unduly prevents or delays the admission of relevant scientific evidence. United States v. Sample, 378 F. Supp. 44, 53 (E.D. Pa. 1974); McCormick, Evidence § 203, pp. 490-491 (2d ed. 1972); cf. United States v. Baller, 519 F. 2d 463, 466 (4th Cir. 1975), cert. denied, 423 U. S. 1019, 96 S. Ct. 456, 46 L.Ed.2d (1975). There are, however, compelling reasons which justify the Frye principle.

Fairness to a litigant would seem to require that before the results of a scientific process can be used against him, he is entitled to a scientific judgment on the reliability of that process.⁷ As stated by Judge McGowan, speaking for the court in United States v. Addison, 498 F. 2d 741, 743-744 (D.C. Cir. 1974):

“[T]he Frye standard retards somewhat the admission of proof based on new methods of scientific investigation by requiring that they attain sufficient currency and status to gain the general acceptance of the relevant scientific community. This is not to say, however, that the Frye standard exacts an unwarranted cost. The requirement of general acceptance in the scientific community assures that those most qualified to assess the general validity of a scientific method will have the determinative voice.”

This is an especially significant consideration with regard to those scientific techniques in which highly subjective judgments are based upon the data received from sophisticated mechanical devices. In these circumstances, the apparent objectivity of the machine may suggest a degree of certainty inconsistent with the subjective aspects of the enterprise.⁸ United States v. Addison, supra, 498 F. 2d at 744; People v. Kelly, supra. As the Supreme Court of California stated in Kelly (549 P. 2d at 1245):

. . Frye was deliberately intended to interpose a substantial obstacle to the unrestrained admission of evidence based upon new scientific principles____ Several reasons founded in logic and common sense support a posture of judicial caution in this area. Lay jurors tend to give considerable weight to ‘scientific’ evidence when presented by ‘experts’ with impressive credentials. We have acknowledged the existence of a'... misleading aura of certainty which often envelops a new scientific process, obscuring its currently experimental nature.’ (Huntingdon v. Crowley, supra, 64 Cal.2d at p. 656, 51 Cal.Rptr. at p. 262, 414 P.2d at p. 390;.....) As stated in Addison, supra, in the course of rejecting the admissibility of voiceprint testimony, ‘scientific proof may in some instances assume a posture of mystic infallibility in the eyes of a jury —’ (United States v. Addison, supra, 498 F. 2d at p. 744.)”

In addition to the advantage of substituting scientific for lay judgment as to scientific reliability, the court in United States v. Addison, supra, 498 F. 2d at 744, pointed out that the Frye test

“. . . protects prosecution and defense alike by assuring that a minimal reserve of experts exists who can critically examine the validity of a scientific determination in a particular case____[T]he ability to produce rebuttal experts, equally conversant with the mechanics and methods of a particular technique, may prove to be essential.”

The dissenting opinion, however, suggests that instead we adopt the rule enunciated by McCormick, that “[a]ny relevant conclusions which are supported by a qualified expert witness should be received unless there are other reasons for exclusion.” McCormick on Evidence § 203 at 491 (2d ed. 1972). McCormick, in opposition to the great weight of judicial authority, believes that disagreement in the scientific community regarding the reliability of a scientific process should go to the weight rather than the admissibility of scientific evidence.

This view seems to us unacceptable. It fails to recognize that laymen should not on a case by case basis resolve a dispute in the scientific community concerning the validity of a new scientific technique. When the positions of the contending factions are fixed in the scientific community, it is evident that controversies will be resolved only by further scientific analysis, studies and experiments. Juries and judges, however, cannot experiment. If a judge or jurors have no foundation, either in their experience or in the accepted principles of scientists, on which they might base an informed judgment, they will be left to follow their fancy.⁹ Thus, courts should be properly reluctant to resolve the disputes of science. “It is not for the law to experiment but for science to do so,” State v. Cary, supra, 99 N.J. Super. at 332.

Nonetheless, under the McCormick standard, juries would be compelled to make determinations regarding the validity of experimental or novel scientific techniques. As a result, one jury might decide that a particular scientific process is reliable, while another jury might find that the identical process is not. However, the reliability of the underlying technique or process to perform as it is supposed to does not vary with different cases. Using the polygraph as an example, although particular polygraph tests may give different results under different circumstances, the basic validity of the polygraph technique in general to give the type of results which are claimed for it does not change with the facts of each ease. Nevertheless, if the trier of facts is to determine the validity of the polygraph test on a case by case basis, one judge or jury might determine that it is reliable and convict or acquit a defendant on the basis of the test results, whereas the very next judge or jury, sitting in the same courthouse and listening to the same operator giving the same type of test results, might determine that the technique is unreliable and ignore the results. Such inconsistency concerning the validity of a given scientific technique or process would be intolerable. See Commonwealth v. Sullivan, 146 Mass. 142, 145 (1888) (Holmes, J.).

Under the Frye test, however, this difficulty is largely avoided. As long as the scientific community remains significantly divided, results of controversial techniques will not be admitted, and all defendants will face the same burdens. If, on the other hand, a novel scientific process does achieve general acceptance in the scientific community, there will likely be as little dispute over its reliability as there is now concerning other areas of forensic science which have been deemed admissible under the Frye standard, such as blood tests, ballistics tests, etc.

In addition, there is a related danger under the McCormick view. The introduction of evidence based on a scientific process, not yet generally accepted in the scientific community, is likely to distract the fact finder from its central concern, namely the rendition of a judgment on the merits of the litigation. Without the Frye test or something similar, the reliability of an experimental scientific technique is likely to become a central issue in each trial in which it is introduced, as long as there remains serious disagreement in the scientific community over its reliability. Again and again, the examination and cross-examination of expert witnesses will be as protracted and time-consuming as it was at the trial in the instant case, and proceedings may well degenerate into trials of the technique itself.¹⁰ The Frye test is designéd to forestall this difficulty as well. As stated in State v. Cary, supra, 99 N.J. Super. at 332:

“All scientific aids and devices go through an experimental and testing stage, and during these stages there may be considerable scientific controversy. During this period of controversy . . . the danger is that a trial may actually result in the trial of the technique rather than the trial of the issues involved in the case, if some less exacting rule is substituted for the time-honored rule of general scientific acceptance,...”

For the foregoing reasons, we agree with the “general acceptance” rule which the Frye case sets forth.

Our adoption of the Frye standard does not, of course, disturb the traditional discretion of the trial judge with respect to the admissibility of expert testimony. Frye sets forth only a legal standard which governs the trial judge’s determination of a threshold issue. Cf. Badman v. Harold, supra, 279 Md. at 169. Testimony based on a technique which is found to have gained “general acceptance in the scientific community” mayhe admitted into evidence, but only if a trial judge also determines in the exercise of his discretion, as he must in all other instances of expert testimony, that the proposed testimony will be helpful to the jury, that the expert is properly qualified, etc. Obviously, however, if a technique does not meet the Frye standard, a trial judge will have no occasion to reach these further issues.

Turning to the admissibility of testimony based on the voiceprint process, prior to 1972 it was generally agreed that the voiceprint process had not been sufficiently tested and accepted to qualify its results for use in the courts.¹¹ The Technical Committee on Speech Communication of the Acoustical Society of America had requested six scientists in the field of acoustics to evaluate the voiceprint process. These scientists, Richard Bolt, Franklin Cooper, Edward David, Peter Denes, James Pickett and Kenneth Stevens, reported in 1970 that the voiceprint process was still in its experimental stage, and the reliability of the conclusions based on the data obtained from the process was uncertain {Speaker Identification by Speech Spectrograms: A Scientists’ View of Its Reliability for Legal Purposes, 47 J. Acoustical Soc’y Am. 597, 603 (1970)):

“[T]he available results are inadequate to establish the reliability of voice identification by spectrograms. We believe this conclusion is shared by most scientists who are knowledgeable about speech; hence, many of them are deeply concerned about the use of spectrographic evidence in the courts.”

In 1971 and 1972, Dr. Tosi and his associates published a series of papers concerning the results of an experiment conducted on the voiceprint process.¹² Subsequently, some courts, relying exclusively on the testimony of Dr. Tosi and his Michigan associates, admitted in evidence testimony based on the voiceprint process. See State ex rel. Trimble v. Hedman, 291 Minn. 442, 192 N.W.2d 432 (1971); Worley v. State, 263 So. 2d 613 (Dist. Ct. App. Fla. 1972) (use for corroboration); Alea v. State, 265 So. 2d 96 (Dist. Ct. App. Fla. 1972) (following Worley)) Hodo v. Superior Court, 30 Cal.App.3d 778, 106 Cal. Rptr. 547 (1973). However, as observed by Judge Kaplan, dissenting in Commonwealth v. Lykus, 367 Mass. 191, 327 N.E.2d 671, 680 (1975):

“It can fairly be said, however, that when the cases were decided the scientific community had not had sufficient time to study Dr. Tosi’s work and reach conclusions as to its possible advance over the previous work in the field. See People v. Law, 40 Cal.App.3d 69, 81-82, 114 Cal.Rptr. 708 (1974). The decisions thus reflected less a consensus in the relevant scientific community that the Tosi method was acceptable, than an absence of study on which an informed opinion could be based one way or the other.”

In 1973, Bolt, Cooper, David, Denes, Pickett and Stevens again addressed the voiceprint issue, in light of the Tosi experiment.¹³ The authors expressed their concern about certain aspects of the Tosi experiment. They mentioned the Tosi experiment’s failure to consider the problems of mimicking or disguising of voices, changes in voice levels, and changes due to stress or other emotional states of the speaker. They expressed further concern over the increase in error rates in comparing voice samples taken at different times, as well as the increase of error in other circumstances. The authors concluded, Bolt, et al., Speaker Identification by Speech Spectrograms: Some Further Observations, 54 J. Acoustical Soc’y Am. 531, 583-534 (1973):

“The Tosi study has improved our understanding of some of the problems of voice identification from spectrograms by indicating the influence of several important variables on the accuracy of identification. In uncovering factors that tend to increase identification errors, however, the study has not given us a definitive answer to the question: ‘How reliably can a person be identified by examining the spectrographic patterns of his speech sounds?’ Under certain laboratory conditions and for some selected sample of the population, the probability of making an error in identification can be stated. But for the less-than-ideal conditions encountered in forensic situations, the indications are that the probability of error will increase substantially. Further studies are needed, with particular attention to the examiner’s decision criteria, the selection of speaker population, the time lapse between voice samples, background-noise conditions, and the psychological condition of the speaker.

“As scientists rather than lawyers, we offer no judgment as to whether or to what extent speech spectrograms should be used for identification in the courts. We wish only to point out that present methods for such use lack an adequate scientific basis for estimating reliability in many practical situations and that laboratory evaluations of these methods show increasing errors as the conditions for evaluation move toward real-life situations. We hope that our explanations of some of the factors that affect speaker identification will provide the legal profession with helpful information on which to base its own judgments concerning the admissibility of the spectrographic method.”

The testimony in the instant case indicates that the fundamental division in the scientific community reflected in these articles has continued without substantial abatement. On direct examination of Dr. Tosi, he acknowledged the division in the scientific community concerning the validity of the voiceprint process:

“Q. How many experts within the field of sound spectrography that have used that process for voice identification oppose that process, who have done actual work in that field?

“A. In addition to the six authorities of Bolt, et al., and none of them worked in voice identification, Stevens had some nine years ago a small experiment. There are three others that oppose it that have done some work — not too much. Some of them have no professional basis. Let’s say five of them, to the best of my knowledge.

“Q. How many of the experts within the field of sound spectrography for voice identification are in favor of that process?

“A. At least — I can give the names of at least 15, and among them very prominent scientists.” ¹⁴

Later, asked about the division of experts, excluding those professionally engaged in the field of voice identification, Tosi testified:

“A. Okay, five were opposed from this reduced group of persons that I said at least have published or done something but were not professionally engaged in the field. I say it is a rough number. Persons that I know of that have done some experimentation or have published, I said less than ten; five opposed — four or five are in favor____

“Q. You are not including Dr. Bolt and his group, are you?

“A. No, I am not.”

Additional expert witnesses who testified for the State, in the instant case, were Sgt. Smrkovski, Dr. Peter Jansen and Dr. John A. McClung. Their testimony was consistent with that of Dr. Tosi in acknowledging the division in the scientific community.

Dr. Donald Baker, an expert witness called by the defense,¹⁵ testified that spectrography was neither a reliable process nor generally accepted within the scientific community. Dr. Baker cited two samplings of opinion of the scientific community, both of which had been unfavorable toward the validity of the process. The first was a meeting in which the members of the Speech Communications Section of the Acoustical Society of America voted 42-0 against the efficacy of the procedure. The second was a mail survey, as reported in a scientific journal, which resultéd in an unfavorable reaction. Dr. Baker also noted that the majority of articles on the subject were negative in their characterization of the process.

The extent of disagreement in the scientific community was emphasized in the instant case by the testimony of Dr. Henry Hollien, another expert witness for the defense,¹⁶ who stated:

“I have conducted or directed about six major studies using [the voiceprint] technique.

“One of the things we have done, and we are the only people who have done this ... we have applied our technique to ... [simulated crimes], and it of course doesn’t work.

“There was nothing wrong with trying to use it [the technique]. It failed. Now it is an abuse.

“Q. It is an abuse because you feel there are some people not qualified to use it?

“A. No, no. It is the data. See, I don’t think the people that use it know about the research literature____There are many studies that have been published which show the problems with this. There is a huge literature that would demonstrate why they should back off, put a moratorium on this until we have some knowledge, and not foster this upon the judicial system and law enforcement agencies. It amounts to a fraud. I don’t think they realize it. They don’t know what is going on, you see.” (Emphasis added.)

There has been a sharp division among the cases which have considered the admissibility of voiceprint evidence after the emergence of the controversy over Tosi’s claims.

Three state supreme courts, California, Michigan and Pennsylvania, have held the evidence inadmissible. Commonwealth v. Topa, supra; People v. Kelly supra; People v. Tobey, supra. In addition, the District of Columbia Court of Appeals, in Brown v. United States, 384 A. 2d 647 (D.C. 1978), has also held voiceprint evidence inadmissible. On the other hand, the Supreme Court of Massachusetts has, in a divided opinion, held the evidence admissible, Commonwealth v. Lykus, supra. And see State v. Williams, 388 A. 2d 500 (Me. 1978).

Two lower state courts have recently ruled voiceprint evidence admissible: People v. Rogers, 86 Misc. 2d 868, 385 N.Y. Supp. 2d 228 (1976); and State v. Olderman, 44 Ohio App.2d 130, 336 N.E.2d 442 (1975). However, in D’Arc v. D’Arc, 157 N.J. Super. 553, 385 A. 2d 278 (1978), the New Jersey Superior Court ruled voiceprint evidence inadmissible.

In the United States Courts of Appeal, voiceprint evidence has been held inadmissible in United States v. Addison, supra, and admissible in United States v. Baller, supra, and United States v. Franks, 511 F. 2d 25 (6th Cir. 1975), cert. denied, 422 U. S. 1042, 95 S. Ct. 2654, 45 L.Ed.2d 693 (1975).

All cases holding voiceprint evidence inadmissible have done so on the ground that the process fails to satisfy the standard articulated in United States v. Frye, supra. It is important to note, however, that neither United States v. Baller, supra, nor United States v. Franks, supra, in holding voiceprint evidence admissible, seemed to apply the Frye test. In Franks, the court stated (511 F. 2d at 33):

“Although we, of course, are aware of the differences of... scientific opinion concerning the use of voiceprints, we also are mindful of ‘a considerable area of discretion on the part of the trial judge in admitting or refusing to admit’ evidence based on scientific processes.” (Emphasis supplied.)

Similarly, in United States v. Baller, supra, the court, after considering the Frye standard, stated (519 F. 2d at 466):

“Unless an exaggerated popular opinion of the accuracy of a particular technique makes its use prejudicial or likely to mislead the jury, it is better to admit relevant scientific evidence in the same manner as other expert testimony and allow its weight to be attacked by cross-examination and refutation.”

The Massachusetts Supreme Court held that voiceprint analysis did satisfy the Frye standard. In Commonwealth v. Lykus, supra, 327 N.E.2d at 678 n. 6, that court stated:

“[W]e agree that there certainly is not uniform and total acceptance of the [voiceprint] method [in the scientific community]____Yet the ... Frye standard does not require unanimity of view, only general acceptance; a degree of scientific divergence of view is inevitable. In this case we are disposed to give greater weight to those experts who have had direct and empirical experience in the field of spectrography____

“Thus, we find the evidence presented in support of the reliability of voiceprints, particularly as expressed in Dr. Tosi’s study, sufficiently persuasive to outweigh the criticism expressed by certain other scientists in the field of acoustics.”

Nevertheless, it is not fully clear whether the Massachusetts court was, consistent with the Frye standard, deciding that the voiceprint method is generally accepted by the scientific community or whether it was attempting itself to determine the merits of the claims of the various scientists. In any event, we find ourselves more in agreement with Judge Kaplan, dissenting in Commonwealth v. Lykus, supra, 327 N.E.2d at 682, who stated:

“To sum up, opinion is divided on the Tosi method; the journal material shows turbulence and discord rather than that ‘general acceptance’ which the Frye case lays down as a precondition of admissibility. Nor can it be plausibly said that those with adverse views are either unqualified to have opinions worthy of respect or are strangers to the relevent scientific ‘field.’ ”

Furthermore, we disagree with the Massachusetts court’s characterization of the nature of the dispute. A degree of scientific divergence of opinion is indeed inevitable, but the degree of divergence surrounding the voiceprint process is fundamental and goes to the very validity of the process itself. This kind and degree of divergence is notably absent in other areas of scientific evidence generally deemed admissible. As stated in Comment, The Voiceprint Dilemma: Should Voices Be Seen and Not Heard?, 35 Md. L. Rev. 267, 280 n. 79 (1975):

“[EJxperts may disagree as to the application of a technique, or as to the results of that application, but they do not generally question that the technique is capable of producing the results claimed. For instance, it is common knowledge that psychiatric diagnoses are often at odds with each other, and it is easy to picture experts disputing whether two writing samples came from the same hand. It is much more difficult to imagine experts disputing whether psychiatric diagnoses or handwriting identifications are possible with any significant degree of reliability. Yet that is precisely the nature of the voiceprint dispute; experts question the capability of the process itself, not just the results of its application.” (Emphasis supplied.)

In addition, it is the almost unanimous opinion in recent legal commentaries that the voiceprint technique does not satisfy the standards articulated in Frye v. United States. See, e.g., Comment, The Voiceprint Dilemma: Should Voices Be Seen and Not Heard? supra; Comment, Voiceprints: The End of the Yellow Brick Road, 8 U.S.F. L. Rev. 702 (1974); Jones, Danger — Voiceprints Ahead, 11 Am. Crim. L. Rev. 549 (1973); Jones, Evidence Vel Non The Non Sense of Voiceprint Identification, 62 Ky. L. J. 301 (1974); Note, Voiceprint Identification, 61 Geo. L. J. 703 (1973); Thomas, Voiceprint — Myth or Miracle (The Eyes Have It), 3 U. San Fern. Y. L. Rev. 15 (1974). Even those authors who advocate the admissibility of voiceprint evidence appear to concede that it does not meet the Frye test and argue instead for alternative revised standards under which it might be admissible. See, e.g., Decker and Handler, Voiceprint Identification Evidence — Out of the Frye Pan and into Admissibility, 26 Am. U. L. Rev. 314, 361-365 (1977); Greene, Voiceprint Identification: The Case in Favor of Admissibility, 13 Am. Crim. L. Rev. 171, 195-197 (1975).

Despite this array, the trial court in the instant case determined that spectrography had achieved the standard of acceptance needed for admissibility. However, the trial court, in holding voiceprint evidence admissible, construed the Frye test to require

“general acceptance ... within the group actually engaged in the use of this technique and in the experimentation with this technique----[W]e are restricting the relevant field of experts to those who are knowledgeable, directly knowledgeable through work, utilization of the techniques, experimentation and so forth, that we are not taking the broad general scientific community of speech and hearing science. In that broad community there probably is not acceptance.”

We have serious doubts that voiceprint analysis meets even this reduced standard. Tosi’s own testimony indicates substantial division of opinion among those who have done work or performed experiments relating to the voiceprint process.

In any event, we find that the trial court’s formulation is inconsistent with the proper standard of acceptance necessary for admissibility. The circumstances of the instant case suggest no basis for “restricting the relevant field of experts” to those who have performed voiceprint experiments, and eliminating from consideration the opinions of those scientists in the fields of speech and hearing, as well as related fields, who, by training and education, are competent to make professional judgments concerning experiments undertaken by others. The purpose of the Frye test is defeated by an approach which allows a court to ignore the informed opinions of a substantial segment of the scientific community which stands in opposition to the process in question.

Thus, based on our examination of the record in the instant case, the judicial opinions which have considered this question, and the available legal and scientific commentaries, we do not believe that “voiceprint” analysis has achieved the general acceptance in the scientific community, at this time, which is required under Frye. We therefore hold that testimony based on “voiceprints” or spectrograms is, for the present, inadmissible in Maryland courts as evidence of voice identification. This holding is, of course, subject to reconsideration by this Court if the use of spectrograms or some other technique of voice identification does in the future achieve the general acceptance of the scientific and legal communities.

Judgment of the Court of Special Appeals reversed, and case remanded to that court with directions to reverse the judgment of the Circuit Court for Montgomery County and remand the case for a new trial.

Montgomery County to pay costs.

1 . No comparison was attempted on the remaining three calls. One call was too short, one too distorted, and no voice exemplar had been secured for a third.

2 . Our grant of certiorari also encompassed Reed’s alternate argument that the Best Evidence Rule was violated when the trial court permitted either a second or a third-hand copy of the tapes of the September 1974 calls to be used for comparison with Reed’s voice exemplars. Reed maintains that the original tapes were destroyed through gross negligence of the investigating police officers, and that the second copy was lost or destroyed without explanation. However, in light of our ruling on Reed’s principal claim, it is unnecessary for us to reach this alternate issue.

3 . See, e.g., United States v. Baller, 519 F. 2d 463 (4th Cir. 1975), cert. denied, 423 U. S. 1019, 96 S. Ct. 456, 46 L.Ed.2d 391 (1975); United States v. Franks, 511 F. 2d 25 (6th Cir. 1975), cert. denied, 422 U. S. 1042, 95 S. Ct. 2654, 45 L.Ed.2d 693 (1975); United States v. Addison, 498 F. 2d 741 (D.C. Cir. 1974); People v. Kelly, 17 Cal. 3d 24, 130 Cal. Rptr. 144, 549 P. 2d 1240 (1976); Commonwealth v. Lykus, 367 Mass. 191, 327 N.E.2d 671 (1975); State ex rel. Trimble v. Hedman, 291 Minn. 442, 192 N.W.2d 432 (1971); Commonwealth v. Topa, 471 Pa. 223, 369 A. 2d 1277 (1977). See also Comment, The Voiceprint Dilemma: Should Voices Be Seen and Not Heard?, 35 Md. L. Rev. 267 (1975); Decker and Handler, Voiceprint Identifícation Evidence — Out of the Frye Pan and into Admissibility, 26 Am. U. L. Rev. 314 (1977); Greene, Voiceprint Identification: The Case in Favor of Admissibility, 13 Am. Crim. L. Rev. 171 (1975); Jones, Danger— Voiceprints Ahead, 11 Am. Crim. L. Rev. 549 (1973); Jones, Evidence Vel Non The Non Sense of Voiceprint Identification, 62 Ky. L. J. 301 (1974); Kamine, The Voiceprint Technique: Its Structure and Reliability, 6 San Diego L. Rev. 213 (1969); Hollien and McGlone, The Effect of Disguise on “Voiceprint" Identification, 2 J. of Crim. Def. 117 (1976).

4 . Dr. Tosi is Professor of Audiology and Speech Sciences and Physics at Michigan State University. He holds two Ph.D.’s, one in Audiology and Speech Sciences from Ohio State University, the other in Engineering and Physics from Buenos Aires University. He is a member of various professional societies and has published several books and numerous papers.

5 . Sergeant Smrkovski has been employed by the Michigan Department of State Police for ten years, the last four and one-half as officer in charge of the voice identification unit. At the time of the hearing, he was about to receive his bachelor’s degree in audiology and speech science from Michigan State University.

6 . See, in addition, Wigmore, The Science of Judicial Proof % 220, p. 450 (3d ed. 1937):

“But, since the additions thus made possible to our unaided senses are due to the use of instruments constructed on knowledge of scientific laws, it is plain that the correctness of the data thus obtainable must depend upon the correctness of the instrument in construction and the ability of the technical witness to use it. Hence, the following three fundamental propositions apply to testimony based on the use of all such instruments:

A. The type of apparatus purporting to be constructed on scientific principles must be accepted as dependable for the proposed purpose by the profession concerned in that branch of science or its related art. This can be evidenced by qualified expert testimony; or, if notorious, it will be judicially noticed by the judge without evidence.

B. The particular apparatus used by the witness must be one constructed according to an accepted type and must be in good condition for accurate work. This may be evidenced by a qualified expert.

C. The witness using the apparatus as the source of his testimony must be one qualified for its use by training and experience.

“These fundamentals will in the ordinary Case be evidenced readily. But in every branch of science there are charlatans and incompetents. The recognition of the above fundamentals will usually serve to diminish the risk of unreliable testimony.”

7 . In 1665, in what may be one of the first reported instances of expert testimony, a certain Dr. Brown of Norwich, testifying at a trial, delivered of himself the expert scientific opinion that the accused were witches and, by practicing their witchcraft at the devil’s bidding, had bewitched several children. The accused were found guilty and hanged. A Trial of Witches at Bury St. Edmonds, 6 Howell’s State Trials 687, 697 (1665). No issue seems to have been raised in that case concerning the validity of the process for determining whether one was a witch.

8 . See, e.g., Highleyman, The Deceptive Certainty of the “Lie Detector”, 10 Hastings L. J. 47. 63 (1958): “[T]he use of ‘lie detector’ evidence invites confusion between (l) the reliability of the objective physiological facts which are recorded by the polygraph, and (2) the reliability of the subjective inferences of truth or deception which are drawn from those facts by the examiner.”

9 . See 2 Wigmore, Evidence § 659, p. 771 (1940):

“When the testimony, thus appearing to the ordinary layman to lack a rational basis, is founded on observations made with esoteric methods or apparatus — vacuum-rays, telepathy, and the like — this method should be explained by the witness; ana, if it is vouched for as accepted in his branch of learning, it suffices to admit his testimony.” (Emphasis supplied.)

10 . See, e.g., the dissent’s observations in United States v. Wright, 17 U.S.C.M.A. 183, 194, 37 C.M.R. 447 (1967):

“The trial was virtually concerned with nothing else but the efficacy and infallibility of the voiceprint process. The court’s questions were directed almost completely to its effectiveness and demonstrated the members’ extreme interest in its identification of the accused, to the exclusion of all others. Moreover, the trial counsel made the process the focal point of his argument, devoting approximately sixty percent thereof to its reliability.”

11 . See State v. Cary, 56 N. J. 16, 264 A. 2d 209 (1970); People v. King, 266 Cal.App.2d 437, 72 Cal. Rptr. 478 (1968). But see United States v. Wright, 17 U.S.C.M.A. 183 (1967).

12 . O. Tosi, H. Oyer, W. Lashbrook, C. Pedrey, and J. Nicol, Voice Identification through Acoustic Spectrography, Speech and Hearing Sci. Lab., Michigan State Univ., Rep. No. 171 (1971); O. Tosi, H. Oyer, W. Lashbrook, C. Pedrey, J. Nicol, and E. Nash, An Experiment on Voice Identification: Excerpts from Report SHSLR 171, Dep’t of Audiology and Speech Sciences, Michigan State University, East Lansing, Michigan (July 1971); O. Tosi, H. Oyer, W. Lashbrook, C. Pedrey, J. Nicol, and E. Nash, Experiment on Voice Identification, 51 J. Acoustical Soc’y of Am. 2030 (1972).

13 . One of the authors, Kenneth Stevens, was apparently also a consultant in the Tosi experiment, Tosi, et al., Experiment on voice Identification, supra, at 2043.

14 . Tosi included in this group of supporters Dr. Peter Ladefoged. Ladefoged was originally an opponent of the voiceprint process (see Ladefoged and Vanderslice, The “Voiceprint" Mystique, 7 Working Papers in Phonetics 126 (1967)). Ladefoged testified in United States v. Raymond, 337 F. Supp. 641 (D.D.C. 1972), where voiceprint testimony was admitted. This result was overturned in United States v. Addison, 498 F. 2d 741, 745 (1974), where Judge McGowan, speaking for the court, observed that:

“[V]iewed in its entirety, Dr. Ladefoged’s letter, as he himself characterized it . . . simply reflects a position ‘of abatement of skepticism towards voiceprint,' not one of complete acceptance.”

In addition, in Jones, Evidence Vel Non The Non Sense of Voiceprint Identification, 62 Ky. L. J. 301, 322 n. 96, Ladefoged is reported as responding to the question, “Would you say that ‘voiceprints’ as a method of voice identification now has general acceptance in the scientific community?” as follows:

“I think I did say that in some case; probably in the Washington case [i.e., United States v. Raymond] I said that. I think now I was in error to say that because, having said that, numerous of my friends, have said, ‘No, not true.’ I said it in good faith thinking that my friends had accepted it, and I now find that I have been reprimanded by some people.

15 . Dr. Baker, who received his Ph.D. in Hearing and Speech Science from Ohio State University, is presently teaching graduate and undergraduate courses in hearing and speech at the University of Maryland. Dr. Baker testified that he had done research in related areas of spectrography and was familiar with the scientific literature concerning voice identification by spectrography.

16 . Dr. Hollien holds advanced degrees including a Ph.D. from the University of Iowa. He is a professor of speech at the University of Florida and is director of the Institute for Advanced Study of Human Communication. He is the head of the Institute’s research program on speaker identification. He is a member of various societies, including the Academy for the Forensic Application of the Communications Sciences, of which he is a council member. He has authored over 100 major publications.