Method of matching a digitized stream of audio signals to a known audio recording

摘要

A system and method matches a digitized stream of audio signals to a known audio recording by receiving an audio stream from an application. The method performs signal analysis on sound and silence in the audio stream to compute a stream audio print of the audio stream, to identify a pattern associated with the audio stream. The method compares the stream audio print to one or more known audio prints from a library of known audio prints. The method first applies a silence constraint to match periods of silence. If application of the silence constraint is successful, the method computes a total scoring metric function to identify a match between the stream audio print and the known audio print. The total scoring metric function employs a scoring metric function that is computed at a subset of temporal points in the non-silent portion of the audio stream audio print.

主权项

1. A method of matching a digitized stream of audio signals to a known audio recording, the method comprising:
receiving, at a processor, an audio stream from an application; performing, at the processor, signal analysis on the audio stream to compute a stream audio print of the audio stream, wherein the audio stream is comprised of sound and silence, wherein the stream audio print is computed on sound and silence, wherein the stream audio print identifies a pattern associated with the audio stream, wherein the pattern is comprised of at least two of sound, silence and frequency, wherein the pattern matches a similar pattern of at least two of sound, silence and frequency in known audio print from a library of known audio prints, and wherein the signal analysis is performed on sound and silence within the audio stream; applying, at the processor a silence constraint to match periods of silence in the stream audio print to periods of silence in the known audio print from the library of known audio prints, wherein the silence constraint is associated with the known audio print; if application of the silence constraint is successful, computing, at the processor, a total scoring metric function associated with a potential match to identify a match between the stream audio print and the known audio print, wherein computing the total scoring metric function comprises:
overlaying the stream audio print with the known audio print;computing a scoring metric function at a subset of temporal points, wherein computing the scoring metric function comprises:
computing the scoring metric function, at a temporal point, between component data from the stream audio print and component data from the known audio print, wherein the component data from the stream audio print comprises:
i) a first frequency associated with the stream audio print;ii) a second frequency associated with the stream audio print; andiii) a first amplitude ratio associated with the first frequency and the second frequency; andwherein the component data from the known audio print comprises:
i) a third frequency associated with the known audio print;ii) a fourth frequency associated with the known audio print; andiii) a second amplitude ratio associated with the third frequency and the fourth frequency; andsumming a result of the scoring metric function at each of the subset of temporal points to create a total score metric, wherein the total score metric is a result of the total scoring metric function; and determining, at the processor, the potential match between the stream audio print and the known audio print.