|
AUDIO ENHANCEMENT
By William T. Andrews, Forensic Audio Analysis, Benton,
Kansas.
Audio recording of interviews, telephone
conversations, and undercover operations is common in law enforcement
today. In addition, video taping of interviews has gained popularity.
These recordings can be used in Court if they meet the rules of
evidence. What are the Uniform Rules of Evidence guiding the admission
and use of the audio or video tapes?
Rule 1002. REQUIREMENT OF ORIGINAL.
To prove the content of a writing, recording, or
photograph, the original writing, recording, or photograph is required,
except as otherwise provided by these rules or by [rules adopted by the
Supreme Court of this state or by] statute.
Rule 1003. ADMISSIBILITY OF DUPLICATES.
A duplicate is admissible to the same extent as an
original unless
(1) a genuine question is raised as to the
authenticity or continuing effectiveness of the original or
(2) in the circumstances it would be unfair to
admit the duplicate in lieu of Rule 1004.
ADMISSIBILITY OF OTHER EVIDENCE OF CONTENTS.
The original is not required and other evidence of the contents of a
writing, recording, or photograph is admissible if:
(1) Originals lost or destroyed.
All originals are lost or have been destroyed, unless the proponent lost
or destroyed them in bad faith.
(2) Originals not obtainable.
No original can be
obtained by any available judicial process or procedures.
(3) Original in possession of opponent.
At a time when an original was under the
control of the party against whom offered, he was put on notice, by the
pleadings or otherwise, that the contents would be a subject of proof at
the hearing, and he does not produce the original at the hearing, or
(4) Collateral matters.
The writing, recording or
photograph is not closely related to a controlling issue.
In addition to the Uniform Rules, Steve M. Solomon,
Jr. vs. Edgar (1955) 92 GA pp207,88 SE2ndl67 and United States vs.
McKeever (1958. DC NY) F Supp 426
set forth the "seven prong predicate" for audio recordings. The
formulation is as follows:
1)
That the recording device was capable of taping the conversation
now offered in evidence.
2)
That the operator of the device was competent to operate the
device.
3)
That the recording is authentic and correct.
4)
That the recording has been preserved in a manner that is shown
to the court.
5)
That the speakers are identifiable.
6)
That the conversation elicited was made voluntarily and in good
faith, without any kind of inducement.
To these rules the following have been added:
1)
That the tape is audible and understandable.
2)
That transcripts may be used in conjunction with the audio
playback to enhance
understanding.
3)
That enhancements of the original recording be admitted, after
proper foundation, in order to assist the trier of fact.
The major problem with audio or videotapes is the intelligibility of the
words spoken. Several major departments such as Houston, Dallas, Los
Angeles, New York, and Miami have units within the department that can
enhance the tapes. Most states have a unit within their State Bureau of
Identification that can enhance audiotapes.
The Federal Bureau of Investigation has an audio laboratory in Quantico,
Virginia that provides authentication, voice identification, transcripts
and enhancements. The major
problem with the FBI is that they receive over 650 audiotapes per month
and the backlog is enormous leading to delays up to several months.
What can the smaller departments do to improve
audio recordings that they have made?
First is the actual recording. Make sure of the
equipment operation and that it is in working order. This may sound
simple but I have received several tapes where the equipment has either
failed resulting in speed fluctuations or was operated incorrectly. If
recording wireless UHF or FM transmissions, make several test recordings
to be familiar with the range, antenna alignment, recorder inputs and
operation. Nothing is worse than to have monitored a perfect signal and
found that the output of the receiver has been patched into the output
of the recorder, giving you a useless tape. When attaching a body
recorder, pay attention to clothing that may constantly rub against the
microphone that will cause the signal of the voice to be masked by the
rubbing. Coach your Cl to be aware of the microphone limitations. They
need to watch out for excess noise and being too far from the subject of
interest. When recording by telephone, be sure to use a telephone
interface or use a recorder such as the Marantz PMD series that has a
built-in modular connector. Watch your volume control so you do not
saturate the tape.
When recording interviews, there are several pitfalls to avoid. Be sure
of your placement of the microphone. It does no good to just record the
questions and not the answers. I have had a tape from a police
department that had interviewed a rape victim at the scene. They used a
Sony M-740 microcassette recorder with a built-in microphone. Sadly, the
microphone was aligned towards the officer and recorded only his
comments. You could not hear or understand the description of the
suspect or vehicle from the victim.
One other problem is the physical room where interviews are conducted.
Try to avoid large concrete rooms with sparse furniture and no carpet.
The echo problem is difficult to overcome. Remember you have binaural
hearing and use both ears and a brain. You can reject sound such as the
echo and concentrate on the original voice. When you have a single
microphone picking up the sound, you have a monaural hearing that cannot
separate the sound. I have had tapes that have had so much echo that the
answers of the person being interviewed were unintelligible.
Second, if your department has less than ten tapes
per year that need enhancement, it would be better to ship them off to a
private lab for enhancement.
Third, if your department has over 1000 tapes per year, it may be
advisable to invest in a stand alone enhancement system. There are
several on the market. Cedar, from England,
has a system using a computer they furnish, and hardware. The cost of
that system is around $32,000. DAC, Digital Audio Corporation, has
several models ranging from $3.000 to $23,000. DAC will provide full
training for law enforcement at their facility in Raleigh, North
Carolina for a small fee.
The next solution for departments with a medium use of tape is to use a
PC with a software program that is developed by several manufacturers. A
good program will have Broad Band Filter, Impulse Noise Filter, Low Pass
Filter, High Pass Filter, Notch Filter, Equalizer with selectable
frequencies, and a limiting compressor. One type of filtering that is
very useful is the "adaptive" filter. This filter changes with the
change of the
noise where as a fixed filter does not. Diamond Cut has a program called
"Live" that costs around $950. Sound Cleaner Pro was developed by the
KGB and is quite powerful for around $900. DiGiCleen is another program
that is new to the market and costs around $700. If you decide to
purchase a software program, be sure to check what is required for the
computer and soundcard to operate the program. Each software program
operates differently and has instructions unique to each program. I
would suggest a Pentium III with at least 800 MHz speed, 512 RAM, 20 gig
hard drive, duplex
soundcard such as Sound Blaster Platinum Windows 98 or 2000 operating
system, a 19" monitor, a color printer to document the filter settings,
and a screen capture program.
The final set-up for enhancement is an analog system. This will require
a small mixer (line and microphone inputs), a graphic or parametric
equalizer, limiting compressor, and an amplifier for small speakers to
monitor your progress. All of this equipment can be purchased for under
$1,500.
Now that you have selected your equipment and it is working, here are
a few tips to get you started in the right direction.
Telephone frequencies are from 330 Hz-3,500 Hz.
Human voice range is from 80 Hz to 50,OOOHz.
Offending hum is normally at 60 Hz.
Offending buzz is normally at 120 Hz.
Tape hiss is Broad Band noise and requires that filter.
Microcassette recorders are at 120 Hz to 5,000 Hz.
Human hearing is from 20 Hz to 20,000 Hz.
It is my hope that this information will get you started on making
better quality audio recordings. You will not have as many case problems
and your conviction rate will increase. If you wish any further
information on enhancement, authentication, voice ID, or transcript
verification please call. Fax, or send me an email at:
Forensic Audio Analysis
William T. Andrews
4936 N.W. County Line Road
Benton, KS 67017
Voice: (316) 799-2770
Fax: (316)799-1021
Email:
wandrew9@earthlink.net
About the author:
William Andrews is a Sustaining Associate Member of the
IAI and the KDIAI. His company, Forensic Audio Analysis,
has been in business since 1990. He has been qualified in
Kansas courts, several other state courts. Federal courts, and
International courts as an expert as well as a voice
identification expert. He is currently the vice chair of the
American Board of Recorded Evidence. He has received the
Robert D. Olson, Jr. award for the best article in 1996 and
1998. He has given educational talks about forensic audio
nationally as well as at the 1996 KDIAI conference.
|
