Anyone claiming to be able to make a voice identification of a disguised voice can be effectively impeached when testifying.
It stands to reason that changing one's voice propulsion has an immediate effect on the frequencies of that voice. Imagine a piano keyboard: each key represents a specific frequency along with its overtones. If one articulates a simple syllable or word such as "oh" in a normal fashion, a number of frequencies equivalent to certain keys would sound. If the voice is disguised when uttering the same syllable, however, a number of other strings would vibrate, not necessarily the same as the original strings.
Understanding the Methodology
In Voice Identification processes, various words and syllables of an "unknown" person are transformed into their respective frequencies by a process known as Fast Fourier transformation. This results in a spectrogram showing the frequencies in varying degrees of black, with the principal frequencies appearing darker than the subsidiary ones. When the test subject repeats the same words and syllables, the resulting frequencies are compared with the unknown's and a decision made. If a number of the frequencies match, then a positive identification can be made.
When a voice is disguised, the frequencies change radically dependent upon the quality of the disguise. For example, if a male voice decides to speak in a hoarse fashion and slower than usual, the frequencies may appear lower and the architecture of the syllables will change considerably. This makes identifying disguised voices incredibly difficult, even following the methods above. Studies done at various colleges indicated that disguised voices cannot be identified.
Cross-Examining the Identification Witness
In cross-examining a witness who has made an identification, questions about the witness's hearing, and whether or not the witness is familiar with and has made spectrograms for positive identification, should establish that nothing scientific has actually proven the validity of the identification.
Should the prosecution bring up concerns about spectography's value as a normal baseline identification method, you can show them a list of more than one hundred companies who use spectrograms for identification and dictation purposes, including IBM and Hewlett Packard. The Air Force is also using spectrograms to identify pilots during air combat so that they can issue instructions to U.S. pilots. Consequently, attempts by the prosecution to denigrate spectrograms should fall by the wayside.
Federal Rule of Evidence 702
Recall that under Federal Rule of Evidence 702, a qualified expert may only testify if certain conditions are established, including that the expert's specialized knowledge will assist the trier of fact, the testimony is based on sufficient facts or data, the testimony is the product of reliable principles and methods, and the expert has reliably applied the principles and methods to the facts of the instant case. As the Notes to the rule indicate, there are many different factors a court will consider in determining whether expert testimony is sufficiently reliable.
Given the difficulties in matching a disguised voice to a normal one, even with the advanced scientific techniques outlined above, if you follow the above suggestions you may be able to effectively discredit any identification of a disguised voice.