Advancing audio forensics of gunshot acoustics.

As submitted by the proposer: This is a scientific study of audio forensic gunshot acoustics and the effects of common digital voice recorders on these high intensity sonic events. More and more justice proceedings involve audio forensic evidence in general, and gunshot audio recordings in particular, due to the steady increase in the number of law enforcement officers who carry personal audio recording devices while on duty, and the nearly 100,000 law enforcement incidents each year involving firearms in the United States (FBI 2012 NIBRS report). Two major issues are addressed. First, the acoustical characteristics of gunshots are currently little understood in an objective sense, and often subject to unscientific physical misunderstandings and subjective assumptions in the forensic field. Second, the response of common personal audio recorders to intense acoustical sounds such as gunshots has not been studied in any systematic manner, and this lack of knowledge seriously limits the ability of an audio forensic examiner to use such acoustic evidence in an investigation. The proposed research involves two phases. In Phase 1, we create the apparatus and methodology for scientific and repeatable collection of firearm acoustical properties, including the important direction-dependence of each firearms sound field. Preliminary work has been published to demonstrate the feasibility of this research. Gunshot acoustical data will be collected for a wide range of firearms using procedures to mitigate sound reflections and reverberation. For Phase 2 we focus upon personal audio recorders typically used by law enforcement officers. We use the gunshot recording methodology developed in Phase 1 to evaluate the extent to which useful forensic information can be obtained from (a) gunshot recordings obtained at sufficient distance from the firearm that the signal does not clip the recorder, and (b) gunshot recordings from near the shooting position that have clipped, degraded, and distorted waveforms due to overloaded recorder electronics. The deliverables for this project include the Phase 1 results: a proposed standard for cataloging firearm acoustical characteristics, and a database of acoustical signatures as a function of azimuth for at least ten distinct firearms. The Phase 2 deliverables are: a systematic description of the strengths and weaknesses of personal audio recorders for close and distant gunshot recordings, and guidelines for audio forensic examiners in dealing with gunshot acoustical evidence from these non-scientific recording devices. Results are published in scientific journals and trade publications, and the methodology and audio database will be provided to NIJ for dissemination. ca/ncf