Archive for October, 2010
Audio authentication and voice identification requires that a forensic expert examine three critical aspects of an audio recording before beginning any forensic process. Whether it be analogue or digital audio recording, an audio forensic expert should inspect the consistent characteristics of the sound wave formations; listen critically to various tones present in the recording, background noise (noise floor) of the audio recording; and examine the electronic spectrograph measurement. These three critical aspects of an audio recording must be consistent throughout the recording to determine authenticity.
An audio forensic expert has been trained by examining hundreds, if not thousands, of hours of audio recordings. This experience helps the forensic examiner to develop a critical listening skill far more precise than the average person’s. That keen sense of sound perception is very important for audio authentication and voice identification.
During the examination process, regardless of analogue or digital audio examination, it is advantageous that the original recording, and recorder, as well as other recording equipment (wireless transmitter, microphone) also be examined. That way, the forensic examiner can recreate the characteristics of the audio recording including signatures (stop-start) and noise floor.
The noise floor is a critical aspect in audio authentication as well as audio identification because it provides the forensic examiner a second dimension of sound to examine and authenticate other than the main recorded signals (speech, gunshot and voice mail).
Alterations in an audio recording, analogue or digital, most likely will be first detected by a change in the noise floor of the audio recording followed by an anomaly that can be heard auditorially, measured electronically and viewed on the computer screen by examining the wave form.
Part of this noise floor is the background noise of the recording. It is the sounds present on the audio recording that the author had not necessarily intended to have recorded but is still part of the recording that is helpful to a forensic examiner.
Both analogue and digital audio recordings have background ambient noise, the noise floor, when the speech or other audio recorded is not present. This background noise speaks volumes on whether the audio recording being examined is original, authentic or has been altered or edited in addition to the examination outcome of the main recorded signals.
All recordings–both digital and analogue–have a noise floor. The term originated when manufacturers of analogue audio recorders referred to the extraneous noise that their machine created in addition to the desired recorded audio signal.
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. These noises can often be removed by the audio forensic expert to help determine facts about the series of recorded events.
Background noise and noise floor extraneous sound can consist of a heating or air conditioning fan running, refrigerator motor, window fan, clock, fluorescent lighting, wind, rain, car running 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, this background noise can interfere with the forensic examination. Clarification is part of the forensic examiner’s job. It is appropriate for the forensic examiner 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 circumstances of each case should be considered by the forensic examiner before any decision has been made by either party.
If the forensic examiner observes characteristics that are noticeably questionable, then the expert must notify the officials in charge of their findings during the preliminary examination phase of the forensic investigation. Original recordings are required, and if not produced, a motion to suppress the evidence should be filed.
Equalization settings can be adjusted to help the forensic expert recover lost or poorly recorded audio. This blog post covers basic audio clarification.
The sound spectrum is measured in frequencies. All audio sounds like a car horn, gunshot and spoken words have frequencies associated with them. Music has the most broad frequency range and covers 80HZ to 12,000 Khz in most cases. In an attempt to keep this blog post easy to understand, I am going to focus on spoken word audio.
Forensic spoken word recordings range in frequencies from 1Khz to 3Khz. Most sound spectrum frequencies above 4Khz and below 800 Hz do not contribute to the spectrum of spoken word recordings. The forensic expert should remove frequencies above and below 1Khz to 3Khz when background noise is present to increase the clarity of spoken word recordings.
Beautiful spoken word fidelity can include frequencies from 400 Hz to 5,000 Khz when recorded in an audio recording studio using state-of-the-art digital audio recording instruments like mixing council, pre amplifiers, microphones and digital computer recording software. However, when it comes to forensic audio, most recordings that need authentication, clarification and identification have been made on digital pocket recorders and wireless interception devices.
These recordings are subject to extreme background noise because they are recorded outside the professional studio environment, next to noisy furnaces and air conditioners and in public environments like restaurants and other places where unwanted chatter and background noise can spoil a recording. Equalization helps the forensic expert begin the clarification and restoration process and is a very important element of restoring the audio recording.
The best way to determine what frequencies need to be adjusted is to first listen to the audio recording. The trained critical listening skills of the audio expert can identify both wanted and unwanted frequencies during the initial listening pass.
Next, the expert will remove frequencies where unwanted sounds exist. Background hums in room ambience, air conditioners and furnace blower motors exist in low range frequency spectrum’s like 250 Hz all the way down to as low as 60Hz.
Radio frequency interference is probably the most difficult distortion to remove from the audio recording. When radio frequencies have static, it is best to first try noise reduction before setting equalization’s settings to clarify the spoken word recordings. The best methods are those that are tried and tried again until the desired outcome is achieved.
It is best for the expert to save the original audio file before applying any equalization treatments just in case it becomes necessary to go back and start over. Also, every step of equalization should be noted in the expert’s work notes so they can recreate the process for the court should it become necessary.
Audio forensic experts have more voice identification cases today than ever before. Legal cases have more audio evidence than ever before and voice identification has become a critical component to many legal proceedings. Voice mail, confidential informant recordings, call verification centers, 911 and other digital audio recordings often time require confirmation of the identity of person speaking.
There are times when the accused is not the person speaking in the recording and it is the audio forensic expert’s job to scientifically prove this conclusion.
The first step in conducting voice identification is for the audio forensic expert to understand how the audio evidence recording was created. The next step is to create an exemplar using as close to the exact same recording methods as possible that were used to create the original to create the exemplar.
An audio exemplar is a recreated audio recording of the person’s voice that is thought to be speaking on the evidence recording. The exemplar is created using as exact recording tools as the original was created with. If the evidence recording was a telephone intercept, the audio forensic expert would use a telephone intercept to record the exemplar. Of course other considerations will come into play as the audio forensic expert determines necessary to maintain the integrity of the voice identification process.
Scientific standards written by the Audio Engineering Society as well as the American College of Forensic Examiners International are referred to by the audio forensic expert when conducting the voice identification. Those standards can be found on the websites of both organizations.
An expert witness is a person who by virtue of education, training, skill, or experience, is believed to have expertise and specialized knowledge in a particular subject. His or her expertise is beyond that of the average person, sufficient that others may officially and legally rely upon the witness’s specialized (scientific, technical or other) opinion about their area of expertise. The expert witness will render an opinion (also known as the expert opinion) about the forensic evidence presented.
Expert witnesses help lawyers develop strategies on how the evidence in question should be argued or presented in court. Often time’s audio evidence may not be original or genuine and should be closely examined by the audio forensic expert to authenticate the evidence or strategize with the lawyer on how to write the motion to have the case dismissed or audio evidence removed from the case.
Expert witnesses have a complete understanding of legal proceedings and can be a huge aid to the plaintiff, defendant and courts when a piece of audio evidence is part of the legal case. Defendants who are facing serious charges especially need to seek the guidance and advice of an audio forensic expert. Plaintiffs who want to establish a solid case should seek the help of an audio forensic expert to authenticate their audio evidence prior to entering the court room. Expert witnesses are valuable resources to courts and other legal professionals.
The first step when restoring audio is to remove background noise. More often than not, a recording in need of sound clarification or restoration has background noise covering the sounds that are desired to be heard. Noise reduction is the process of reducing and often eliminating that unwanted background sound. Sound like wind, motors, lawnmowers, electronic hums and buzzes and other sounds that may be louder than the spoken word and cover that speech so it is not audible.
Once the background noise is removed the desired spoken words can be heard. There are many levels of clarification acceptance. Some of the time the recording needs to be clarified so a jury can hear it. Other times the recording may only need to be audible to a transcriptionist or court reporter. Then, once the transcription is created, the audio forensic expert can go back, listen to the clarified recording comparing to the transcript and correcting any discrepancies using the expert’s critical listening skills.
The recording can then be certified by the forensic expert and an affidavit created as to the genuineness of the transcript for the legal proceeding. That way, if the audio recorded evidence is difficult to hear in a court room, the audio expert not only certifies the audio recorded evidence but can also testify on its accuracy.
This is a good example of a strategy between lawyers and expert witnesses. It is the expert witness’s job to suggest strategies to lawyers, public defenders, police and other government agencies that retain the expert witness. The lawyer will have difficulty and not represent the client to the best of their ability if they do not understand the audio forensic process and retain an audio forensic expert. The audio forensic expert of course cannot develop a strategy without a lawyer who is licensed to present the law case in the court room.
Before beginning your case that includes audio evidence, consult with a qualified audio forensic expert.
Yep It’s Him, Thomas Pynchon
By Steven Kurutz
Last Tuesday, long-suffering fans of the reclusive writer Thomas Pynchon received a double gift. Pynchon’s latest book, “Inherent Vice,” a stoned-out detective story set in early-‘70s L.A., was released by Penguin Press (read the Journal’s review). And to promote it, the publisher put out a cool video trailer featuring a narrator whose slow, lazy cadence sounds suspiciously like that of Pynchon’s, as evidenced by a guest appearance on “The Simpsons” and this clip from what appears to be a German TV spot. Inquiries by GalleyCat and others as to whether Pynchon is the guy channeling the novel’s main character, beach bum private eye Doc Sportello, have been met with “no comment” from Penguin Press and the video’s producers, Meerkat Media. And, of course, the man himself is mum (Would Pynchon fans expect anything else?).
In an effort to solve the mystery, Speakeasy did a little sleuthing and called Ed Primeau, a Michigan-based sound engineer and voice identification expert. Like handwriting analysis, voice identification is an inexact science, often used by law enforcement to rule out a suspect rather than to provide a 100% clear-cut ID. Still, people have unique vocal timbres and deliveries, especially Pynchon, who sounds like actor John Astin (i.e. Gomez Addams from the old TV show), mixed with a Midwest corn farmer, with a dollop of aging stoner.
So is it possible to rule out the man in the “Inherent Vice” trailer as being the same guy in the Simpsons episode and German TV clip? Not at all, according to Primeau. In fact, he says, based on a preliminary analysis the speech pattern and inflection is “virtually identical” in all three clips. “It’s a very unique style of delivery,” Primeau says. “It’s very up-and-down. He’ll hit these accented spots every few words. You know the TV show “Dragnet,” how Joe Friday talked? It’s the opposite of that.”
We should point out Primeau is an unbiased witness, having never read Pynchon (“I don’t know this guy but it looks like he has some history as an author,” he said). Nevertheless, if he hasn’t been taken by the man’s work, Primeau is intrigued by his voice, which he describes as “a tobacco-driven soft rasp.”
Primeau’s conclusion: “Beyond a reasonable degree of professional certainty, I believe these voices were delivered by the same person.” Confronted with Primeau’s findings, Tracy Locke, a publicist at Penguin, came clean and admitted, “It is, in fact, Thomas Pynchon doing the narration.”