Methods and apparatus for generating audio fingerprints for calls using power spectral density values

Abstract

The present invention relates to methods, systems, and apparatus for processing audio signals. An exemplary method embodiment includes the steps of: removing silence from an audio signal; determining, for a plurality of time segments of the audio signal, power spectral density values of the audio signal for each of a plurality of N different frequency bins, N being an integer greater than 1; identifying (i) a plurality of dominant frequency peaks based on the determined power spectral density values, and (ii) positions in the audio signal corresponding to the identified peaks; and generating a first audio fingerprint from at least some of the identified plurality of dominant frequency peaks and the identified positions in the audio signal corresponding to the identified peaks. In various embodiments, audio fingerprints are generated from an audio signal of call and then used to determine if the call is a robocall or SPAM call.