PEOPLE
v.
TOBEY
Docket Nos. 18301, 18302.
Michigan Court of Appeals.
Decided April 23, 1975.Frank J. Kelley, Attorney General, Robert A. Derengoski, Solicitor General, William F. Delhey, Prosecuting Attorney, and Lynwood E. Noah, Assistant Prosecuting Attorney, for the people.
O'Brien, Moran & Dimond, for defendant.
Before: DANHOF, P.J., and McGREGOR and D.F. WALSH, JJ.
Leave to appeal applied for.
*422 McGREGOR, J.
The defendant was convicted of two counts of illegal sale of heroin by a jury, on August 8, 1973, and sentenced to two concurrent terms of 10 to 20 years in prison.
On February 17, 1972, Officer VanTiem purchased 1/4 ounce of heroin for $200 from the defendant. Officer VanTiem had been introduced to the defendant through Allen Lang and confidential informant I.L. 333, neither of whom testified at trial. On February 28, 1972, Officer VanTiem recorded two telephone conversations with a person he believed to be the defendant. The next day, VanTiem went to the defendant's apartment and purchased one ounce of heroin for $900.
Following the defendant's arrest, the trial judge granted the prosecutor's motion to compel the defendant to make voice exemplars into a recording device, using the identical words spoken in the telephone recordings, to enable the prosecution's experts to run a voiceprint test on the two tapes.
At trial, the prosecution called three expert witnesses on the subject of voiceprint identification: Dr. Oscar Tosi,[1] Lt. Ernest Nash[2] and Police *423 Officer Lonnie Leonard Smrkovski. Based on Dr. Tosi's testimony, the trial court ruled, over defendant's objection, that evidence of voiceprint identification tests made under proper conditions was admissible.
Lt. Nash testified that a voiceprint analysis had been run on the tape of the telephone recording made on February 28th, and the tape made by the defendant under court order. He stated that, in his opinion, the two voices were from the same person. The trial court then admitted the two tapes with a limiting instruction and allowed them to be placed before the jury.
Officer Smrkovski testified that he had conducted the voiceprint analysis on the two tapes. Based on this analysis, Officer Smrkovski opined that it was the defendant who had made the incriminating remarks in the telephone conversation with Officer VanTiem.
Defendant appeals as of right from the jury's verdict of guilty on both offenses.
Defendant first claims that the trial court's instruction on the defense of entrapment was erroneous, since the instruction included the use of the subjective test. In People v Turner, 390 Mich 7; 210 NW2d 336 (1973), the subjective test for the defense of entrapment was rejected. Instead, the Court adopted the objective theory which stresses the conduct of the police and not the predisposition of the defendant. However, the trial in the case at bar was concluded one month prior to the Supreme Court's decision in Turner, supra. Since Turner has prospective force only, the trial court did not err by allowing the jury to consider the *424 defendant's predisposition to commit the offense. People v Koehler, 54 Mich App 624; 221 NW2d 398 (1974), People v Sinclair, 387 Mich 91; 194 NW2d 878 (1972).
The defendant next claims that he was denied a fair trial because of the failure of the prosecution to disclose the identity of the government informant. We find this issue without merit.
Paid Informant I.L. 333's only activity in this case was to introduce Officer VanTiem to Allen Lang and to the defendant. There is no testimony upon which this Court could infer any other involvement of this informant after the initial contact with the defendant.
In People v Henley, 54 Mich App 463; 221 NW2d 218 (1974), the defendant raised this identical issue on appeal. The Court held there that, since the informant was at no time present during the events leading up to the time of the heroin delivery, the state was not compelled to present the informant as a res gestae witness. Therefore, the refusal to disclose the identity of I.L. 333 was not erroneous.
Defendant next contends that the trial court erred by denying his motion to sever the two offenses for separate trials. We disagree.
The prosecution claims that the two offenses, which occurred 12 days apart, constitute a single transaction for purposes of double jeopardy and, therefore, must be tried together. People v White, 390 Mich 245; 212 NW2d 222 (1973).
In People v Martinez, 58 Mich App 693; 228 NW2d 523 (1975), the defendant took the same position on appeal as the prosecution in the case at bar. The Court, in Martinez, rejected this position and stated as follows:
*425 "The deliveries of heroin in the instant case were made to the same agent during the course of a continuous undercover investigation. But these facts alone do not relate the events intimately enough so as to characterize them as being a part of a single transaction under the test adopted in People v White. Nine days separated the two sales in the instant case; the amounts involved were substantially different; and the record does not disclose any connection between them, such as an agreement after the first delivery to return for another sale.
"We hold that the two deliveries therefore constituted separate transactions and that the defendant's plea-based conviction for the second delivery did not place him twice in jeopardy. People v White, supra."
Defendant claims that he has the right to sever, for separate trials, an information which charges in separate counts the commission of two distinct offenses of the same nature. He is mistaken.
In Pointer v United States, 151 US 396; 14 S Ct 410; 38 L Ed 208 (1894), the defendant was charged in separate counts with two murders and, although the crimes were allegedly committed by the defendant on the same day and in the same county and district, it did not affirmatively appear from the indictment that they were the result of one transaction, or that they were "connected together". The indictment did, however, show upon its face that the two offenses were of the same class or grade of crimes, and subject to the same punishment.
The defendant objected to the joinder in the one indictment of the separate counts of murder, claiming that the prosecutor should be required to elect upon which one of the charges he would go to trial.
On appeal, the U.S. Supreme Court upheld the trial court's denial of the defendant's motion, stating, at 403:
*426 "While recognizing as fundamental the principle that the court must not permit the defendant to be embarrassed in his defense by a multiplicity of charges embraced in one indictment and to be tried by one jury, and while conceding that regularly or usually an indictment should not include more than one felony, the authorities concur in holding that a joinder in one indictment, in separate counts, of different felonies, at least of the same class or grade, and subject to the same punishment, is not necessarily fatal to the indictment upon demurrer or upon motion to quash or on motion in arrest of judgment, and does not, in every case, by reason alone of such joinder, make it the duty of the court, upon motion of the accused, to compel the prosecutor to elect upon what one of the charges he will go to trial. The court is invested with such discretion as enables it to do justice between the government and the accused. If it be discovered at any time during a trial that the substantial rights of the accused may be prejudiced by a submission to the same jury of more than one distinct charge of felony among two or more of the same class, the court, according to the established principles of criminal law, can compel an election by the prosecutor."
In the present case, we cannot say that the trial court erred or abused his discretion in denying the defendant's motion to sever the counts for separate trials. Even had the court granted defendant's motion, the prosecutor could properly have introduced testimony under MCLA 768.27; MSA 28.1050 of the other narcotic sale.
We must emphasize that this case does not involve the situation where several offenses are charged, distinct in point of law, and the trial of these several offenses would involve proof of substantially different transactions, thereby confusing the defendant in his defense or depriving him of substantial rights.
This distinction was made in People v McKinney, *427 10 Mich 54 (1862), where the defendant was charged with multiple counts of embezzlement arising out of eight different conversations between August and December of 1860. The defendant claimed the court erred in failing to quash or compel the prosecution to make an election as to which count should go to the jury.
On appeal, the Court found no error and concluded, at 95-96;
"In the present case the information charges apparently several offenses of the same kind; and if the evidence related to several substantially different and distinct transactions, it would have been a proper case for putting the prosecutor to his election."
Since the defendant in the instant case was not embarrassed in his defense by the joinder of the two counts of the same offense in a single trial, the trial court did not abuse its discretion in denying the defendant's motion to sever. MCLA 767.75; MSA 28.1015.
The defendant's last two assignments of error concern the prosecution's use of the voiceprint identification technique. Defendant first contends that the trial court's order compelling him to furnish voice exemplars for identification purposes through comparison with other voice recordings violated his Fifth Amendment right against self-incrimination.
The Supreme Court has recently decided this issue, in United States v Dionisio, 410 US 1; 93 S Ct 764; 35 L Ed 2d 67 (1973). In Dionisio, the Court held that compelling a grand jury witness to furnish a voice exemplar did not violate the Fifth Amendment privilege against self-incrimination, since the exemplar was to be used solely to measure the physical properties of the witness' voice, *428 and not for the testimonial or communicative content of what was to be said. Therefore, the defendant's right against self-incrimination was not violated when he was compelled to give a voice exemplar into a tape recorder, nor were his constitutional rights violated when the prosecution played that exemplar before the jury.
Lastly, defendant claims that the court erred in admitting voiceprint identification information, since the prosecutor failed to lay a proper foundation for the admission of this testimony.[3] We agree.
Before the evidence of spectrograph[4] analysis can be admitted, the prosecutor must lay a proper foundation of proof of the accuracy of the scientific and mechanical instruments used in the test. The trial court's admission of the spectrograph comparison over the defendant's timely objection, based *429 upon the prosecutor's failure to lay a proper foundation for this evidence, was reversible error.
The foundation laid in this case was insufficient in three ways. First, the samples which were compared for analysis were not similar. The unknown tape was recorded over the telephone, while the compelled exemplar was not. Dr. Tosi testified that direct versus telephonic obtaining of the sample is a variable which must be considered. Further, Officer Nash testified that the telephone eliminates high and low pitch sounds below 300 Hz. and above 3,500 Hz., and admitted that fully 8% of vocal range lies below 300 Hz. No compensation in the spectrograph was made for this variance; neither was the compelled voice exemplar played through a telephone in order to get similar samples for the comparison. We note, however, that Officer Nash did testify that research has indicated that this discrepancy is probably insignificant in voice identification.
Secondly, the spectrograph itself lacked calibration and proper maintenance. Proof of a scientific or mechanical instrument's accuracy and proper use is a requisite for the admission of the results derived from such devices. People v Kenney, 354 Mich 191; 92 NW2d 335 (1958). In the case at bar, the voiceprint operators admitted that the machine used was never checked until "nothing came out". Neither was there a periodic maintenance or calibration of the machine. Thus, it cannot be known whether the spectrograms made here were in any way accurate.
The third and most important defect in the prosecution's foundation was the length of time between the taking of the unknown tape and the taking of the exemplar from the defendant. The unknown tape recording was made on February *430 28, 1972; the trial judge granted the prosecution's motion to compel the defendant to give a voice exemplar on December 6, 1972. No experiment has ever verified the use of samples taken more than one month apart. Further, the Michigan State University Voice Identification Project, performed under the direction of Dr. Tosi, demonstrates that if spectrograms were used from samples made a month apart, more errors were made than using samples made at the same time. The Voiceprint Technique: A Problem In Scientific Evidence, 18 Wayne L Rev 1365 (1972).
Another problem presented in this case is the circumstances surrounding the preservation and custody of the tapes. Officer Smrkovski testified that he was aware that tapes deteriorate over time and in different humidities. Because of this, Officer Smrkovski testified that after August 21, 1972, he kept the surreptitious tapes in a special vault with a relative humidity maintained constantly at 65%. However, prior to that date, the tapes were merely stored in a desk drawer. When offering real evidence, an adequate foundation for admission requires testimony first, that the object offered is the object which was involved in the incident and, further, that the condition of the object is substantially unchanged. People v Beamon, 50 Mich App 395; 213 NW2d 314 (1973). Since the tapes were improperly stored, we cannot say that their condition at trial was substantially unchanged from the time when the recordings were made.
Admission of the spectrograph comparison of the tapes was prejudicial. Viewing the record in its entirety, we cannot say that the jury might not have reached a different result if the voiceprint identification had been properly excluded.
Reversed and remanded for a new trial.
NOTES
[1] Dr. Tosi is Professor of the Department of Audiology and Speech Sciences and Physics at Michigan State University, 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; he has qualified as an expert in these fields in the courts of almost a dozen states.
[2] 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. Since 1968, he has been a student at Michigan State University, 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 heretofore been qualified as an expert in the courts of 20 states. He has examined approximately 180,000 sound spectrograms of the human voice.
[3] Since the prosecution failed to lay a proper foundation for the admission of the results of the spectrograph comparison, we need not decide the issue of whether voiceprint identification is admissible as scientific evidence under the test set forth in Frye v United States, 54 US App DC 46; 293 F 1013 (1923), and People v Morse, 325 Mich 270; 38 NW2d 322 (1949).
[4] The spectrograph has been thus described: The sound spectrograph consists of 4 basic parts: (1) a magnetic recording device, (2) a variable electronic filter, (3) a paper-carrying drum that is coupled to the magnetic recording device, and (4) an electric stylus that marks the paper as the drum rotates. The magnetic recording device is used to record a short sample of speech. The duration of the speech sample corresponds to the time required for one revolution of the drum. Then the speech sample is played repeatedly in order to analyze its spectral contents. For each revolution of the drum, the variable electronic filter passes only a certain band of frequencies, and the energy in the frequency band activates the electric stylus so that a straight line of varying darkness is produced across the paper. The degree of darkness represents the varying amplitude of the speech signal at the specified time within the given frequency band. As the drum revolves, the variable electronic filter moves to higher and higher frequencies, and the electric stylus moves parallel to the axis of the drum. Thus, a pattern of closely-spaced lines is generated on the paper. This pattern, which is the spectrogram, has the dimensions of frequency, time, and amplitude. United States Department of Justice, Voice Identification Research A Summary of the Report to the Law Enforcement Assistance Administration, Grant No. NI-70-004, February 1971, at 9.