Reconsideration of forensic acoustics

The high talent level of in our Society members is unquestioned. However, it appears that some of us have not yet reached full maturity relative to Forensic Acoustics. While the adversarial aspect of Forensics does not parallel our investigational rigor, the number of variables they face is much greater than ours. Accordingly, development of acoustic systems for forensic use requires different approaches than we typically use. Speaker identification (SI) constitutes a good example here. Fifty years ago, it was agreed that a valid SI system could be quickly created but that has not happened. In response, the undersigned adapted the scientific method (and its standards) as a model for SI system development. The resulting research was modestly successful but the approach was not considered “useful”. Later, the Daubert test confirmed this model (but ordinarily was circumvented). Now, the National Academy of Science has verified these models and is insisting on proper developmental protocol. It is no longer pe...