Stay in Control

A Practical Application for Forensic Audio and Video Examinations

By Wayne R. Runion

The term ‘control’ has long meant something specific and invaluable in the world of chemistry. But in the last few years, it’s come to mean something somewhat different – yet just as critical – to the world of forensic audio and video.

Random House Dictionary defines it as “to test or verify by parallel experiment or other standard of comparison.” Controls are widely used in the scientific community to verify results. One familiar example is the use of placebos in drug testing, in which placebos provide a point of reference or comparison for the experiment. But relative to forensic science, ASLCD considers control to mean “a standard of comparison for verifying or checking the finding of an experiment,” and requires a control conducted with each examination and the results documented (paragraph 1.4.2.8 of the ASCLD 2005 manual).

ASLCLD-LAB brought the issue of how controls were to be used in the Digital Evidence community to the attention of the Scientific Working Group on Digital Evidence (SWGDE), of which I am a former member. At the time, ASLCLD-LAB was incorporating the new forensic disciplines of Digital and Multimedia Evidence for accreditation. SWGDE found controls valuable for the forensic audio community and outlined a recommendation that controls be used in forensic audio in the SWGDE Best Practices for Forensic Audio:

“A control is a known audio test signal that is run through a system to ensure that the system produces the expected result. This gives an examiner confidence that the system will perform as expected when processing evidence. In forensic audio, the system is the complete signal path including playback, processing, and recording equipment, cables, and connectors. A control test should be run on the complete signal path before processing evidence.

The interval at which controls are run should be appropriate for the specific equipment used. Specifically, controls should be run:

• Whenever a system configuration change occurs.


• Regularly for equipment that experience wear.


• When infrequently used equipment is put into service.
  

If a control test fails, evidence should not be run through the system until it behaves properly. When failure occurs:

• Notify other examiners of the failure.


• Troubleshoot the system to isolate the failed component. Recognize that the failure could be in a piece of equipment, a cable, a connector, or in an interconnection itself. Take the failed component out of service until it can be replaced, repaired or recalibrated, or it otherwise demonstrates reliable performance.” (chapter 4 version 1.0, January 31, 2008)

The Scientific Working Group on Imaging Technology (SWGIT) also outlines the importance of controls for forensic video examinations in their Best Practices For Video Analysis:

For tape based media
Prior to inserting videotape evidence into a playback device, ensure the equipment is functioning properly by inserting a non-evidentiary test tape of known signal and image quality. When playback of the evidentiary tape is less than optimal or signal dropouts occur, and the analyst suspects player idiosyncrasies as a potential factor, multiple players and/or recorders should be utilized to preview the tape. In some cases, this may necessitate retrieving the original recorder and/or camcorder unit. For example, head misalignment on the original recorder may produce a tape in which video playback is degraded or not viewable when played back on any unit other than the original recording device. Tracking adjustment may be necessary to optimize playback of the original video. (draft January 18, 2008)

In practice, a control is a known sample recording that contains audio and video (i.e. tone and color bars) of a known quality, played through the same equipment as the evidence recording. The control aids the examiner in determining if the equipment is functioning properly and all the connections are correct. My control contains known music and tones, color bars, black frames and the old movie countdown.

Maybe at this point you’re wondering if control is really important beyond being a necessary step for accreditation. Simply put, yes. A control actually serves several purposes. Use the same path or chain of equipment and cables that you intend on using for the analysis of your audio and video evidence. This will check that playback heads and cables are all in good working order and any intermediate devices (such as a compressor/limiter) are also not adversely affecting the audio or video.

For perspective, consider the dangers of not doing a control. Unintentional or accidental noises could be introduced on the digitized evidence tape. This might leave you with the mistaken impression that the signal was very noisy from the source, or just a really poor recording. You could be processing out noise that had accidentally been induced. Cables between the playback and the computer could be bad, or a dirty head or roller on an audio or video cassette player could grab and damage the tape. Running a control before performing an audio or video analysis not only provides information about the signal path, it also may save the evidence from harm.

Even if you are not in a laboratory environment and do not fall under accreditation standards, it is good practice to run a control before each and every examination. This will help ensure that your system and all intermediate devices are working properly, that no accidental noise has been added, and help prevent surprise damage to the evidence by dirty heads. Before you start processing evidence, run a control: it will help ensure that you begin with the best possible digital copy of your audio and video evidence recording, making your resulting products the best they can be.