Extracting an excerpt from a media object
View Patent ↗An excerpt of a media object is extracted by computing, for each bar of an N-bar loop, one or more perceptual quality vectors. For each of the one or more perceptual quality vectors within a search zone (S), one or more distances between bar i and bar i+N is computed and sorted to generate a sorted list of bars.
1. A method for extracting an excerpt of a media object, comprising:
computing, for each bar of an N-bar loop of a media object, one or more perceptual quality vectors, where N is an integer;
computing, for each of the one or more perceptual quality vectors within a search zone (S), one or more distances between bar i and bar i+N, correspondingly, where i is an integer;
sorting the one or more distances between bar i and bar i+N to generate a sorted list of bars; and
extracting an excerpt from the media object based on the sorted one or more distances for at least one of playback and searching for content.
2. The method of claim 1 , wherein the sorting includes the step of removing, for each of the one or more perceptual quality vectors, a predetermined number of bars from the sorted list of bars.
3. The method of claim 1 , wherein the sorting includes the steps of:
weighting the one or more distances by weighting each of the one or more distances between bar i and bar i+N; and
accumulating the weighted distances.
4. The method of claim 1 , further comprising the step of:
detecting a repetition of the N-bar loop within the search zone (S).
5. The method according to claim 1 , further comprising the step of:
providing a crossfade between at least one section of the N-bar loop, wherein the crossfade has a crossfade length (D).
6. The method according to claim 5 , wherein providing the crossfade includes:
prepending the crossfade to the starting section of the N-bar loop at T 0 −D, where T 0 is the time onset of a first beat of the N-bar loop.
7. The method according to claim 1 , wherein the one or more perceptual quality vectors include a pitch vector, a timbre vector and an energy vector.
8. The method according to claim 1 , wherein the search zone (S) is based on a duration of the media object (T D ), a predetermined introduction section (S 1 ), and an outro section (S 2 ), wherein S=T D −S 1 −S 2 .
9. A non-transitory computer-readable medium having stored thereon one or more sequences of instructions for causing one or more processors to perform the methods in any one of claims 1 - 8 .
10. A system comprising:
one or more processors; and
a computer-readable storage medium coupled to the one or more computer processors, the computer-readable storage medium comprising instructions stored thereon, wherein the instructions when executed by the one or more processors cause the one or more processors to perform the method in any one of claims 1 - 8 .