| 摘要 |
Various embodiments are directed to a computer system, computerized method, and computer programs can further comprise the one or more processors being programmed at least to:
sample the new media item to create an audio signature sample;
compare the audio signature sample of the new media item against a signature database to identify at least one matching audio signature for a media item;
identify common portions of the audio signature sample of the new media item and the matching audio signature;
map the common portions of the audio signature sample of the new media item and the matching audio signature to one or more common segments in a segment database, the segment database comprising a plurality of segments corresponding to a segment of the media item;
obtain additional contextual data about the new media item, the contextual data including a video, narrative or both;and
determine if the additional contextual data is substantially similar for each common segment. |
