Profiling hoax callers

Abstract

Hoax calls annually cost law enforcement and security agencies over a billion dollars, and sometimes lives. Bogus bomb threats, “swatting” calls to the police, hoax calls to the coast guard etc. cause these agencies to respond, deploying personnel and resources needlessly. The response itself could cause direct danger to innocent citizens, while also drawing resources away from genuine emergencies that could otherwise have been expeditiously attended to. Law enforcement agencies would hence benefit greatly from technologies that could assist them in identifying the circumstances of a hoax call, or identifying the hoax callers themselves. These could lead to more informed responses to hoax calls, or to the arrest of the perpetrators. In this paper we describe technologies to profile hoax callers. Profiling in this context refers to the estimation of the speaker's personal traits, and of their physical surroundings, from their voice. This is a difficult task, particularly because the hoax calls are often very short, degraded in audio quality, and highly dramatized, with the callers attempting to disguise their voices to prevent identification. We present aspects of current technology in various fields that we have applied to this problem, and the challenges that remain in formulating reliable solutions.