As an audio forensic expert who conducts voice identification, I receive calls from people around the world asking about voice identification. The American Board of Recorded Evidence provides voice identification standards that help me determine if I can identify a piece of audio evidence.
Some of the requests I receive are a stretch because the audio in question does not have enough words spoken in order for me to create an exemplar. An exemplar is one of the most important tools to a voice identification test. The exemplar must be made as closely to the original recording as possible. The audio forensic expert often does not have as much control over the technology when making the exemplar as they would like. In that case careful attention must be paid to the variable electronic readings in spectrograph measurement allowing or compensating for the variable.
In other words, if a piece of audio evidence was a telephone intercept or voice mail recording, the exemplar must be made using a telephone and recording device as similar as possible to the device that was used to create the original recording in question. This is where the attorney representing our side in the litigation comes in. The attorney must petition the court during discovery to help the audio expert learn what equipment was used to record the original evidence.
This process was especially important back in the analogue days. Today’s digital recordings create much higher quality recordings and have different authentication processes.
Here are the American Board of Recorded Evidence voice identification requirements as accepted in the scientific community:
Standards for Comparisons Determination
The following are the standards accepted nationally by all professional organizations involved with voice identification, including the FBI, the Audio Engineering Society, the International Association for Identification, and the American Board of Recorded Evidence:
- IDENTIFICATION: At least 90% of all comparable words must be very similar aurally and spectrally, producing not less than twenty (20) matching words. The voice samples must not be more than six (6) years apart.
- PROBABLE IDENTIFICATION: At least 80% of the comparable words must be very similar aurally and spectrally, producing not less than fifteen (15) matching words.
- POSSIBLE IDENTIFICATION: At least 80% of comparable words must be very similar aurally and spectrally, producing not less than ten (10) matching words.
- INCONCLUSIVE: Falls below either the Possible Identification or Possible Elimination confidence levels and/or the examiner does not believe a meaningful decision is obtainable due to various limiting factors.
- POSSIBLE ELIMINATION: At least 80% of comparable words must be very dissimilar aurally and spectrally, producing not less than ten (10) words that do not match.
- PROBABLE ELIMINATION: At least 80% of the comparable words must be dissimilar aurally and spectrally, producing not less than fifteen (15) words that do not match.
- ELIMINATION: At least 90% of the comparable words must be very dissimilar aurally and spectrally, producing not less than twenty (20) words that do not match.
For more on Voice Identification, check out Ed Primeau’s latest book, “That’s Not My Voice!” available on Amazon.
photo credit: knitronica-87 via photopin (license)
