Audio visual signature, method of deriving a signature, and method of comparing audio-visual data
View Patent ↗The invention relates to the analysis of characteristics of audio and/or video signals for the generation of audio-visual content signatures. To determine an audio signature a region of interest for example of high entropy—is identified in audio signature data. This region of interest is then provided as an audio signature with offset information. A video signature is also provided.
1. A method of comparing audio or audio-visual data in a processor in which a plurality of comparisons are made between first audio signature data derived from two or more audio samples of a first audio or audio-visual sequence and second audio signature data derived at plurality of temporal positions within a second audio or audiovisual sequence and the temporal position of best match of audio signature data is established;
wherein each said audio signature is formed by:
low pass filtering in an infinite impulse response (IIR) filter stage;
high pass filtering in a finite impulse response (FIR) filter stage;
decimating in a decimator; and
taking the sign of the output of the decimator.
2. A method according to claim 1 in which the degree of similarity of the audio signature data is used to derive a measure of the similarity of the first and second audio or audio-visual data sequences.
3. A method according to claim 2 in which the temporal position of best match is compared with timing data associated with the first audio signature data so as to derive a measure of audio timing difference between the first and second audio or audio-visual data sequences.
4. A method according to claim 3 in which a measure of video timing difference is compared with the measure of audio timing difference so as to derive a measure of the difference in relative video to audio delay between the first and second audiovisual data sequences.