Dewey C. HODO, Petitioner, v. SUPERIOR COURT, RIVERSIDE COUNTY, Respondent; PEOPLE of the State of California, Real Party in Interest.
In 1968, People v. King, 266 Cal.App.2d 437, 72 Cal.Rptr. 478, held, based on the record before the court at that time, that evidence of spectrogram or voiceprint identification was inadmissible. However, the court did not close the door on the eventual admissibility of this technique assuming that future events would establish its scientific reliability and acceptance.1
During the ensuing four years scientific research in this field has continued and the technique has received recognition in other jurisdictions. Now, the record before this court indicates that voiceprint identification is scientifically reliable and has gained sufficient acceptance in the scientific community to admit into evidence the opinion of an expert voiceprint reader.
The matter comes before this court on a petition for writ of prohibition following the denial of a motion under Penal Code, s 995.
Petitioner is charged with violation of Penal Code, s 96(1) (promise of a juror to render a verdict for a party), and Penal Code, s 93 (offering as a juror to receive a bribe). At the preliminary hearing evidence was adduced that the defendant was a juror in a condemnation action. The property owner received a telephone call indicating that the caller would like to see him make a lot of money from his property. The authorities were advised of this call and two days later the property owner received another call from the same party which call forms the basis for the pending criminal charges. This call was recorded. An investigation ensued during which the defendant gave police officers a tape recording of his voice. A witness, whose testimony is discussed below, testified over objection that the unknown voice recorded in the telephone conversation and the defendant's recorded voice came from the same person. This evidence established probable cause. No issue is made of the sufficiency of the evidence if this evidence is admissible.
The sole issue is the admissibility of voice identification by the use of spectrographic recordings known as voiceprints.
The two witnesses on this issue were Dr. Oscar Tosi and Detective Lieutenant Ernest Nash.
Dr. Tosi is Professor of the Department of Audiology and Speech Sciences and Physics at Michigan State University. He holds two doctorates, one in Audiology and Speech Sciences and Electronics from Ohio State University, and the other in Engineering and Physics from Buenos Aires University. He is a member of a number of societies in the fields of Speech, Logopedics and Phoniatrics. He has published two books and more than 35 papers in the fields of Audiology and Phonetics.2 He has qualified as an expert in these fields in the courts of ten states.
Dr. Tosi is also an expert in the method of voice identification known as voiceprints. This is a method of voice identification which consists of identifying or eliminating an unknown voice among several known by both listening to the voices and visually inspecting a spectrogram. Acoustical spectrography is a branch of science which consists of composing the voice or the sound into harmonic components and obtaining a visual pattern of the sound which pattern is called a spectrogram.
When Dr. Tosi first entered the field of voiceprint identification in 1966, he could express no opinion as to its reliability. After making preliminary studies in the field, Dr. Tosi, in 1967, expressed the opinion that the method of voiceprint identification showed promise but that he could not either recommend, endorse, or reject the method without more comprehensive investigation. He felt that additional scientific evidence was called for. Thus, Dr. Tosi testified in ‘State v. Cary,’ that he advised the Court in New Jersey not to use the voiceprint identification method since it was his opinion that he did not have enough evidence to prove the validity of the technique.3
However, Dr. Tosi continued with his studies and now is of the opinion that the technique is scientifically reliable and is generally accepted by others who would be expected to be familiar with its use.
Dr. Tosi has conducted special studies in the field of voiceprint identification starting in 1966 when he was teaching spectrography at Michigan State University. At that time, he was contacted by the Michigan State Police re the possibility of using the voiceprint to identify or eliminate suspected persons. Then in 1968, he applied for and received a $300,000.00 grant from the United States Department of Justice to conduct comprehensive experimentation in the field. This he did from 1968 to 1970. 35,000 trials of identification or elimination were conducted as part of this laboratory experiment. In addition, some 700 voices were obtained from actual cases. He selected 250 persons representing a population of 25,000 to conduct his experiments on. He selected a homogeneous population of voices with no speech impairments. All of the persons selected were from a close range of ages, young people, all with the same kind of education and background, all of them speaking the so-called General-American English midwest dialect.
He first recorded sentences in very different contexts and different types of translation. He tested speakers recording directly into a tape recorder or into a telephone. This was done first in a quiet environment and then in a noisy environment. He then had speakers uttering clue words and then other isolated words. These same clue words were put in a fixed context in ongoing normal speech and then a random context, to observe the effect of the use of the clue word system. It was found that the clue word system produces a difference in spectrography.
Dr. Tosi conducted both open and closed trials. In the closed trials the examiner knows that the unknown is among the known. In the open trial the examiner does not know whether the unknown is among the known.
The voiceprint method of identification consists of both listening and visually matching the voices. During the 35,000 trials held the errors of wrong identification came out to be six per cent. The easiest type of trials had only 0.5 per cent error while the most difficult had six per cent misidentification. In analyzing the six per cent error it was determined that this was caused by the fact that Dr. Tosi forced his experimental examiners to reach a positive decision even if they were in doubt.
The examiners had four choices:
1. Almost uncertain.
2. Fairly uncertain.
3. Fairly certain that my decision is right.
4. Almost certain that this is the right decision.
In processing out the results, it was determined when the examiners put down numbers 3 and 4, they produced 2.4 per cent misidentification.
Dr. Tosi accounts for the 2.4 per cent misidentification because the student examiners were allowed only fifteen minutes to complete a trial and the fact that the examiners were never allowed to hear the voice tapes.
However, a professional examiner could have as much time as he deems necessary to complete an identification. In addition, the professional examiner would actually listen to the voice tape. Since the professional examiner is not a naive listener but is trained to listen to a voice this would aid him in making an identification.
Based upon Dr. Tosi's background, experience, studies and his research, he is now of the opinion that when a professional examiner has all the time he needs, is responsible for his decisions, and where he may listen to the voice samples as well as visually examine the voice spectrogram, then the voiceprint method of identification is extremely reliable.4
Based upon Dr. Tosi's knowledge of the vocal track and the mechanics of speech production as well as an analysis of almost 50,000 spectrograms, he has concluded that no two people would create the identical pattern on the voice spectrogram.
As Dr. Tosi completed his testimony, he gave his opinion that Lt. Nash is the best professional examiner using the voiceprint method of identification.
Ernest Nash is a Detective Lieutenant with the Michigan State Police and is the officer in charge of the Voice Identification Unit of the Scientific Crime Laboratory of the Michigan State Police. He has been in charge of that unit since 1967.
He studied originally for some 15 months under Mr. Kersta (see below) and since 1968 has been a student at Michigan State University (presumably under Dr. Tosi) where he has majored in Audiology and Speech Sciences. He has had extensive experience in making sound spectrograms of individual voices as they relate to voice identification. He is a member of several societies in this field and has presented numerous papers on the subject of voiceprint identification. He has qualified as an expert in voiceprint identification in the courts of 12 states. He has examined approximately 80,000 sound spectrograms of the voice of a human being.
Lt. Nash received the two recordings in the instant case, i.e., the recording of the telephone call to the property owner and the tape recording of the interview with the defendant. With each tape recording he prepared sound spectrograms of the voice. Approximately a dozen sound spectrograms were made of the known voice of petitioner as well as the unknown voice of the individual who made the telephone call. He then subjected both recordings to the voiceprint method of identification which is basically the critical listening through a headset to the questioned and the known voices while simultaneously visually examining the patterns of the voiceprints and making a visual examination and comparison. He had ample opportunity to both hear the voice and look at the voiceprint as no deadlines were placed upon him with respect to this examination. He felt free to make as many sound spectrograms as he desired and also felt free to state, if such were the case, that he was unsure of his identification.
Based upon his comparison of the sound spectrogram of the voicepring and listening to the two voices, Lt. Nash was of the opinion that the unknown voice (the telephone call) and the voice identified as the petitioner are one and the same and could not be the voice of any other human being. Of this, Lt. Nash is convinced beyond any doubt. He believes from his study and from his experience regarding voice spectrograms that there are no two people in the world whose voice spectrograms would be identical.
Petitioner contends that voiceprint identification is not accepted generally by members of the scientific community who are familiar with its usage and that the underlying theory is not reliable or supported by scientific and statistical analysis.
The test for the admissibility of scientific testing is whether or not it has received general acceptance by recognized experts in the field. (Huntingdon v. Crowley, 64 Cal.2d 647, 51 Cal.Rptr. 254, 414 P.2d 382; People v. King, Supra, 266 Cal.App.2d 437, 443, 72 Cal.Rptr. 478; People v. Williams, 164 Cal.App.2d Supp. 858, 331 P.2d 251.)
‘The determination of whether a scientific test has received general acceptance by recognized experts in the field so as to justify the admission of expert testimony based on the results of the test is primarily a question of fact for the trial court. (Citations.)
‘The trial court enjoys considerable latitude in determining the qualification of an expert. Its ruling will not be disturbed upon appeal unless a manifest abuse of discretion is shown. (Citations.)
‘The standards used in the exercise of judicial discretion in determining the qualifications of an expert, like other questions of law, are subject to appellate review. (Citation.)
“The opinion of an expert must not only be on a proper subject . . . but must be based on reliable ‘matter’ (‘facts, data, and such matters as a witness' knowledge, experience, and other intangibles upon which an opinion may be based’; . . .)' (Citation.)
“Wigmore states the rule as follows: ‘When the testimony thus appearing to the ordinary layman to lack a rational basis is founded on observations made with esoteric methods or apparatus . . . the method should be explained by the witness, and if It be vouched for as accepted in his branch of learning, it suffices to admit his testimony.” (Italics added.) (Citation.)’ (People v. King, Supra, 266 Cal.App.2d 437, 443, 72 Cal.Rptr. 478, 482.)
‘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.’ (Frye v. United States, 54 App.D.C. 46, 293 F. 1013, 1014.) This is also the rule in California. (See Huntingdon v. Crowley, Supra, 64 Cal.2d 647, 653—654, 51 Cal.Rptr. 254, 414 P.2d 382.)
Based upon the above authorities, the court in King rejected the voiceprint identification as that issue was presented by the evidence in that case.
The foundation for the admissibility of voiceprint identification in that case was based upon the testimony of Mr. Lawrence Kersta. Mr. Kersta is obviously a pioneer in this field and apparently something of a zealot whose enthusiasm for his subject carried his opinion beyond the area of acceptance either in the scientific community or in legal circles. Based upon the record before the court in King, we would have no hesitancy in agreeing with the result reached in that case. However, four years have elapsed since King and further research in the field as related by Dr. Tosi persuades us that the time has now come to accept this type of evidence in courts.
Obviously, some comparison of Mr. Kersta and Dr. Tosi is called for.
Mr. Kersta has a Master's Degree in Electrical Engineering and Physics. Dr. Tosi holds two Doctorates, one in Audiology, Speech Sciences and Electronics, the other in Engineering and Physics. Mr. Kersta's only published work in a scientific journal as of the date of King was an article in Nature Magazine. Dr. Tosi has published two books and 35 papers in science journals. Mr. Kersta's experiments were not in speech identification as were Dr. Tosi's but in speech classification. There was no showing that Mr. Kersta had any knowledge of the functions of the body in producing speech. He was purely an engineer, an employee in the Research Laboratory of Bell Telephone Company. Dr. Tosi does have knowledge of the vocal track and the mechanics of speech production.
The basis for Mr. Kersta's opinion was certain laboratory tests with 123 employees of the Bell Laboratory. Dr. Tosi conducted over 35,000 tests involving over 250 speakers. At the time of Mr. Kersta's testimony in King, voiceprint technology was in its preliminary stage of development. Further scientific testing and experiments by Dr. Tosi have established the voiceprint method of identification and it has now become accepted as a science. As indicated, Mr. Kersta was a pioneer. Dr. Tosi has now carefully, scientifically and objectively verified some of Mr. Kersta's preliminary subjective opinions. According to Dr. Tosi, the scientific community has now changed its attitude about the reliability of voiceprint method of identification and there has been an abatement of skepticism among the scientific community as to the reliability of this technique.
In other words, in 1968, the scientific community, including Dr. Tosi, was not of the opinion that voiceprints were reliable. Mr. Kersta on a narrow statistical base entertained the subjective opinion that they were. Dr. Tosi, after extensive subsequent experiments and research, has now overcome his initial skepticism and now is of the opinion that they are reliable. Based upon the record in this case, we agree.
The petitioner contends that the technique is not generally accepted by those in the field by reason of the following colloquy which occurred during cross-examination of Dr. Tosi:
‘Q. Is the identification of the person who is speaking in an unknown sample, by the technique you have described here and which we expert momentarily Lieutenant Nash to give an opinion, a process which is generally accepted in your field, that is to say, the field of acoustics, the field of linguistics and related sciences?
‘A. Since it is a new process, it is not generally accepted because it is new. It isn't a field known, so it is not like a fingerprint that is accepted with no question. This is a new kind of technique, so it is not sufficiently widespread, so the answer would be no.’
However, Dr. Tosi's subsequent testimony tempered that statement in a very considerable degree.
‘Q. And just to make sure that we understand each other—
‘Q.—I want to ask it a different way: Is it true that this technique is not generally accepted in your specific field of specialty?
‘A. No. I would say that, presently, after our experimentations, more and more is accepted by the scientific community through making aware of exactly the percentage of errors and the kind of validity that this could be assessed to this method, in my opinion.
‘Q. Again, it is true, is it not, that this technique is not generally accepted within your field of specialty? Is that not a true statement?
‘A. No. I would not say that this is absolutely a true statement, because most of the colleagues of my field, they are now reverting (sic) their first opinion and more and more are becoming familiar and acquainted with the method, and they have not presented any kind of objections like they used to do.
‘. . .sen
‘Q.—is this technique generally accepted by your colleagues, the people who are of similar statute to yourself in the field of linguistics and related sciences?
‘A. In all fairness, I should answer that those who really are familiar with spectrography, they are accepting the technique. . . .
‘. . . al
‘Q. I don't know how many people might be in your field. I have no idea, but if I went to every one of them and asked this question: ‘Is this a reliable technique to identify the human voice?’, what percentage of those people would say, ‘Yes,’ in your opinion, and what percentage would be ‘No’?
‘A. Those who are familiar with the technique, those who know the method, those who know the method, who are familiar with the techniques, I would say most of them.
‘Q. And I'm going to ask you, sir, are most of your colleagues in your field of specialty familiar with this technique?
‘A. No, sir.
‘Q. Why not?
‘Q. Well, because they have other fields of interest. They are not interested in this, but since I am not interested in linguistics, myself, so you cannot ask me one opinion on a particular theory of linguistics, because I am not interested, so I am not—therefore, I cannot give an opinion, and my opinion would be not valid, but those persons in my field that I know are familiar with the process of voiceprint, now most of them have reversed their opinion. Now they are much in favor of this system if properly applied by a qualified, trained, qualified professional, if he is entitled to give no opinion when he is in doubt, if he is listening to the voice. . . .
‘. . .if
‘Q. What percentage of these people accept your technique as scientifically valid?
‘A. In my opinion, all those who are familiar with—in my opinion, they all accept this technique presently.’
The solution to some of the seeming inconsistencies in Dr. Tosi's testimony is answered by a statement in People v. Williams, Supra, 164 Cal.App.2d Supp. 858, 331 P.2d 251, in which the court held that the results of the Nalline test were admissible.
‘No experts were called by the defendants and the expert testimony as above summarized stands uncontradicted in the record with this exception: Each of the People's experts did admit on cross-examination that the medical profession generally is unfamiliar with the use of Nalline and therefore it cannot be truthfully said that the Nalline test has met with general acceptance by the medical profession as a whole, general acceptance being at present limited to those few in a specialized field who deal with the narcotic problem.
‘Should this fact render the testimony inadmissible? We believe not. All of the medical testimony points to the reliability of the test. It has been Generally accepted by those who would be expected to be familiar with its use. In this age of specialization more should not be required.’ (Emphasis added.) (People v. Williams, Supra, at pp. 861—862, 331 P.2d at p. 253.)
Thus, the tenor of Dr. Tosi's testimony is that the technique is generally accepted by those experts in the field who would be expected to be familiar with its use.
Dr. Tosi submitted an impressive list of those among his colleagues who now share his present opinion as to the reliability of the voiceprint technique including ‘Dr. Ladefoged,’ presumably Dr. Peter Ladefoged, Professor of Phonetics in charge of the Phonetics Laboratory at UCLA, who testified in the King case that Kersta's tests were unreliable, and ‘Dr. Gerstman,’ presumably the Dr. Gerstman mentioned in King, who testified that Kersta's tests were unreliable. In summary, most of the skeptics have become believers.
Dr. Tosi's research during the last four years as detailed above affords basis for a finding by the trial court that the technique known as voiceprint identification is now supported by scientific and statistical analysis and enjoys the acceptance by recognized experts in the field.
Other jurisdictions which have accepted voiceprint identification are as follows:
(1) In 1967, the United States Military Court of Appeals held voiceprint identification admissible for purposes of corroboration in the case of United States v. Wright, 17 U.S.C.M.A. 183, 37 C.M.R. 447;
(2) In 1971, the Supreme Court of the State of Minnesota in State ex rel. Trimble v. Hedman, 291 Minn. 442, 192 N.W.2d 432, held that voiceprint identification was admissible to corroborate voice identification. The court traced the history of voiceprint identification pointing out the weaknesses of Mr. Kersta's original opinion but holding that Dr. Tosi's subsequent research had established the necessary foundation for the admissibility of the technique. The court noted that experts may disagree but that disagreement alone does not make an opinion inadmissible;
(3) In 1972, the District Court of Appeal of the State of Florida in Worley v. State, 263 So.2d 613, concluded after reviewing the testimony of Dr. Tosi and Lt. Nash that voiceprints were admissible to corroborate a police officer's testimony identifying the defendant as a person that made a false bomb threat by telephone. Again, the court traced the history of voiceprint research from Mr. Kersta through Dr. Tosi. The court noted the cases of State v. Cary, Supra, and People v. King, Supra, and in regard to these cases made the following observation: ‘Both of these cases are over three years old. Since then, impressive scientific data has been amassed as to the voiceprint's reliability. We hypothesize that, based on the changes in Dr. Tosi's testimony alone, Cary and King would be decided differently today. For these reasons we decline to follow these cases.’
(4) In 1972, another division of the District Court of Appeal of the State of Florida in Alea v. State, 265 So.2d 96, again following the expert testimony of Dr. Tosi and Lt. Nash, concluded that voiceprint identification testimony was admissible;
(5) In 1972, the United States District Court in United States v. Raymond, 337 F.Supp. 641, held voiceprint evidence to be admissible. The court observed that spectrography had first gained prominence in the scientific community some ten years previous through the pioneer study of Mr. Kersta but that despite his strong claims, few acoustical scientists shared his confidence. However, the court noted that these reservations were not based on the belief that spectrograms could not produce the results claimed but that Mr. Kersta's experiments had not demonstrated with sufficient certainty the reliability of spectrogram analysis. However, the court felt that the subsequent research by Dr. Tosi had supplied the necessary scientific reliability and remedied the basic defects in Mr. Kersta's studies. In this respect, the court said: ‘The real import of the Tosi study is that it remedies the two major defects of the Kersta study. First, Kersta was criticized for using a Heterogeneous sampling of unknown voices, i.e., the spectrograms used represented speakers with different accents, of different ages and backgrounds, and that this fact made it easier to differentiate between speakers. Tosi, on the other hand, used a Homogeneous sampling of 250 students at Michigan State University each of whom was carefully screened by Tosi's associates from a group of over 25,000 students. Thus, the 250 selected each spoke what is referred to as non-accented, or General-American English, had no noticeable speech defects, were all male, undergraduate students, and ranged in age from 19 to 31. (Fn. omitted.) The second major criticism of the Kersta experiment was that it was conducted using only Closed testing groups, i.e., that the spectrogram of the unknown voice was always included in the group of known voices being used. Thus, in the Kersta test, all the examiner had to do was find the sample in the known group of spectrograms that Most closely matched the spectrogram of the unknown voice in order to make an ‘identification.’ Dr. Tosi, mindful of this defect, set up both open and closed experiments, i.e., in the open tests, the examiners were told that the spectrogram of the unknown voice May or may not be among the spectrograms of the known speakers. (Fn. omitted.) Thus, if the spectrogram of the unknown speaker did not match one of the known speakers' spectrograms, no identification would be made. This ‘open’ situation more closely parallels the actual situation confronted by law enforcement officials when making voice identifications, in that the voice spectrograms of various suspects may or may not be that of the unknown caller.' (United States v. Raymond, Supra, at p. 643.)
Thus, the record in the instant case indicates that since King voiceprint identification has received general acceptance by recognized experts in the field who would be expected to be familiar with its use and has therefore reached the standard of scientific acceptance and reliability necessary for its admissibility into evidence. An impressive area of other jurisdictions has so held. Therefore, we hold, based on the record in the court below, that there was no error in receiving into evidence the testimony of Lt. Nash.
1. ‘To paraphrase the statement of the California Supreme Court in Huntingdon v. Crowley, Supra, (64 Cal.2d 647) at p. 656, 51 Cal.Rptr. (254), at p. 262, 414 P.2d (382), at p. 390, some day in the near or distant future ‘voiceprint’ tests may achieve the same degree of acceptance as the present standard blood tests. ‘In that event the courts will welcome their use as an aid in our never-ending pursuit of truth. “”’' (People v. King, Supra, 266 Cal.App.2d 437 at p. 461, 72 Cal.Rptr. 478, at p. 493.)
2. Audiology, in connection with the speech science, is concerned with the production, transmission, and reception of speech. Phoniatrics is the field which deals with impairment of speech. Logopedics is a part of phoniatrics, but more directed to the teaching processes relating to speech defects and improvements.
3. In State v. Cary, 99 N.J.Super. 323, 239 A.2d 680 (1968), the New Jersey Superior Court stated of Dr. Tosi's testimony, ‘He was of the opinion that the technique has considerable potential as an aid to law enforcement, but before he would give a firm scientific opinion he felt that further experimentation and testing was required because of its infancy in the related scientific fields.’ (State v. Cary, 239 A.2d at p. 683.) In the instant case, as will develop, Dr. Tosi testified that his opinion had changed since Cary because of further research and experimentation.
4. For a more exhaustive explanation of the theory and technique of spectrography or voiceprint identification, see People v. King, Supra, 266 Cal.App.2d 437, 447—449, 72 Cal.Rptr. 478.
GARDNER, Presiding Justice.
KERRIGAN and KAUFMAN, JJ., concur.