Voice Identification Standards: Practical and Official

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.

To learn more about voice identification, call 800.647.4281

Voice Identification Line Up

Recently, I was asked to listen to 23 phone intercept recordings and determine if any of the voices in these conversations were repetitive.  As an audio forensic expert, when I conduct a voice identification test, I have to have an exemplar or voice sample of the accused.

Back to back voice samples are the first step to a task like this.  Each telephone conversation included two voices. The first thing I did was separate the voices and create two new audio project files. That way I can critically listen to all voices back to back in order to determine if any of the voices were identical or at least had similar characteristics. 

These telephone recordings were created by federal law enforcement and were very clean.  No noise reduction was necessary as it was back in the analogue days.

Once I completed this back to back assembly process, I had the recordings transcribed so I could choose phrases and sentences to use when creating the exemplar.  An exemplar is a known sample of speech recorded as exact as possible to the original evidence.  The exemplar is created under supervision so I know the identity of the person speaking (who is the accused).

When the exemplar recording is complete, those phrases that were recorded are now inserted into the original evidence recordings in the new audio project files for critical listening.  In this particular case, I noticed that the exemplar did match some of the telephone conversation evidence that was recorded by the federal authorities. 

The next step is to create work notes listing all the similarities as well as differences observed during the critical listening phase of the voice identification testing.  These notes help me create my report when the voice identification testing is complete.    

I also use spectrum analysis and sonograms to help with the identification process.  I often print out the display of these two electronic measurement devices and include these print outs with my report. 

One thing I have learned over the 25+ years as a forensic expert is to keep it simple.  Judges like an uncomplicated decision from a qualified forensic examiner.  They become frustrated when they have to interpret new information they have never heard of or are not familiar with.

Voice identification is both an art and a science. As a voice identification expert, I use my talent skill and ability in every case I am assigned to.  The science is acceptable in court and the art is the ability to adapt every case to scientific standards.  

Clarification for Authentication of Audio Recordings

The following article was written to help the legal community better understand the audio clarification process. The descriptions it includes will also help a lay person better understand the value of audio clarification.

All digital and analogue recordings have a noise floor. The term “noise floor” originated when manufacturers of analogue audio recorders referred to the extraneous noise that their machines created in addition to the desired recorded audio signals.

Often, a background noise constitutes most of the audio recording and covers a portion of speech that needs to be audible in order to determine a series of events pertinent to the case. Noises like the humming of a fan, air conditioning, heating systems, wind, and other unwanted sounds contribute to the audio recording noise floor. These noises can often be removed by the audio forensic expert (who is also known as the examiner) to help determine facts about the series of recorded events.

The extraneous sounds of background noise and noise floor can consist of a heating or air conditioning fan running, refrigerator motor, window fan, clock, fluorescent lighting, wind, rain, the running engine of a car, and even radio or television. All these sounds contribute to the background noise and noise floor of a recording and aid the forensic examiner in authenticating a recording. However, although these sounds can authenticate the environment of an alleged crime, this background noise can interfere with the forensic examination. It is appropriate and part of the forensic examiners job to remove these background sounds in order to authenticate or clarify an exhibit of audio recorded evidence.

Some of the recordings experts are asked to authenticate are confession recordings created by law enforcement agencies. Defendants exclaim, “That is not what I said, they edited it” or “There is more I said that has been edited out of the recording”. Due process entitles both parties in litigation to examine any evidence presented in their case. However, original recordings are not always available for examination. How do you as a law enforcement official feel about the absence of original recordings?

I have worked on cases where missing “original evidence” was considered spoliation of evidence. Personally, I believe that the circumstances of each case should be considered by the forensic examiner before any decision has been made by either party.

If there are noticeable questionable characteristics that the forensic examiner observes regarding the audio evidence, then the expert must notify the officials in charge of the findings during the preliminary examination phase of the forensic investigation. Original recordings are required for litigation cases. If they are not produced, a motion to suppress the evidence should be filed.

What is an Audio Forensic Expert? Audio Authentication Video by Ed Primeau

Ed Primeau 800-647-4281
If you ever wondered what audio forensic experts do, this video will help you understand the science as well as the art of audio forensics. In this video you will see me discover three anomalies in this under cover audio recording.