Method, system and storage medium for monitoring audio streaming media

摘要

The present disclosure provides method and a computer system for tracking audio streaming media. The method includes: after streaming media information matching a currently played audio stream segment is identified, determining whether the streaming media information matches a next audio stream segment, and when the streaming media information matches the next audio stream segment, directly returning a result; or when the streaming media information does not match the next audio stream segment, performing streaming media information matching again for the next audio stream segment, which occupies less memory, improves a calculation speed, and reduces calculation complexity of matching the streaming media information.

主权项

1. A method for tracking audio streaming media, comprising:
at a computer having a display, one or more processors and memory for storing a plurality of program modules to be executed by the one or more processors:
receiving an audio stream;segmenting the audio stream into multiple audio stream segments including a currently played audio stream segment and a next audio stream segment, wherein the two consecutive segments are separated by a time interval;identifying first streaming media information corresponding to the currently played audio stream segment by using its audio fingerprint, and using the first streaming media information for representing the currently played audio stream segment, the audio fingerprint being a content-based digital signature representing an acoustic feature of the currently played audio stream segment, wherein the identifying step further comprises:
performing framing process on an audio signal of the currently played audio stream segment to acquire multiple framing spectrum diagrams;detecting a key frame among the framing spectrum diagrams by using a start point detection algorithm, and reserving the framing spectrum diagram corresponding to the key frame and discarding the other framing spectrum diagrams;acquiring an audio fingerprint of the key frame from the framing spectrum diagram corresponding to the key frame, and comparing the audio fingerprint of the key frame with pre-stored audio sub-fingerprints to acquire an accurate matching audio sub-fingerprint from the pre-stored audio sub-fingerprints; andidentifying, along a timeline defined by the pre-stored audio sub-fingerprints, streaming media information corresponding to the accurate matching audio sub-fingerprint as the first streaming media information corresponding to the currently played audio stream segment;displaying the first streaming media information on the display of the computer;calculating a Hamming distance between an audio fingerprint of the next audio stream segment and the audio fingerprint of the accurate matching audio sub-fingerprint corresponding to the first streaming media information, so as to obtain a matching degree between the next audio stream segment and the first streaming media information;determining whether the matching degree is greater than a preset threshold;when the matching degree is less than the preset threshold, identifying second streaming media information corresponding to the next audio stream segment according to the step of identifying the first streaming media information corresponding to the currently played audio stream segment;when the matching degree is greater than the preset threshold, choosing the first streaming media information as the second streaming media information; andreplacing the first streaming media information displayed on an interface with the second streaming media information.