| CPC G10L 17/22 (2013.01) [G06Q 10/10 (2013.01); G06Q 50/265 (2013.01)] | 20 Claims |
|
1. An automated method of processing audio information to identify persons of interest in regard to a public-safety incident, the method comprising:
receiving, from an audio-capture device and via an electronic processor of a computer-aided dispatch (CAD) system, a plurality of human-generated narrative statements as audio;
converting, with the electronic processor, the plurality of narrative statements into a plurality of corresponding digital texts;
determining, with the electronic processor, a respective incriminating score for each of the narrative statements by identifying, in a respective one of the corresponding digital texts, incident information classified as not being publicly known;
determining, with the electronic processor, a cumulative score for an individual based on one or more of the respective incriminating scores tagged to the individual;
making, with the electronic processor, a determination of whether the individual is a person of interest in regard to the public-safety incident based on a comparison of the cumulative score with a fixed threshold; and
performing, based on the determination, at least one automated responsive action selected from the group consisting of updating a record in a database including a non-transitory computer-readable medium, tagging a video captured by a camera and associated with the audio, and generating a recommendation for transmission over a communication network to a public-safety officer.
|