Granted Patent B2

US 12,567,445 · App. 18/562,230 · Granted Mar 3, 2026

Inventors: Imran Rashid (London, GB); Jordan Smith (London, GB); Fraser Smith (London, GB); Will Hawkins (London, GB); Edmund Philip Newton-Rex (London, GB)

Assignee: LEMON INC.

G11B27/031

Abstract

Claims (37)

1 . A method for generating a video remix, the method comprising:

receiving an input video;

selecting at least one excerpt from the input video, wherein an audio signal of the selected excerpt includes at least one onset, wherein the at least one excerpt has a predetermined length;

determining a plurality of sub-sequences of the at least one excerpt, wherein an n-th sub-sequence of the plurality of sub-sequences has a length corresponding to an n-th divisor of the predetermined length, wherein the n-th divisor equals to the (n−1)-th power of two, and n is an integer greater than two; and

rearranging the plurality of sub-sequences according to a predetermined pattern to form the video remix, wherein the video remix has a length corresponding to a multiple of the predetermined length.

2 . The method of claim 1 , wherein the at least one excerpt begins at the at least one onset.

3 . The method of claim 1 , wherein the method further includes:

detecting the at least one onset by applying an onset detection function to the input video.

4 . The method of claim 3 , wherein the method further includes:

applying onset selection in cases where more than one onset in the input video is detected, wherein the onset selection is based on at least one of onset strength, onset loudness and onset density.

5 . The method of claim 1 , wherein determining the plurality of sub-sequences includes chopping the excerpt into sub-sequences, wherein a majority of the sub-sequences includes the at least one onset.

6 . The method of claim 1 , wherein determining the plurality of sub-sequences includes determining a loop point corresponding to an onset within the first half of the excerpt, wherein sub-sequences with a shorter length than the excerpt begin at the loop point.

7 . The method of claim 1 , wherein the predetermined pattern corresponds to a musical backing track.

8 . The method of claim 1 , wherein the method further includes:

selecting a second excerpt from the input video such that an audio signal of the second excerpt includes at least one onset;

determining a plurality of sub-sequences of the second excerpt; and

arranging the plurality of sub-sequences of the first and second excerpts into the video remix according to a predetermined multi-clip pattern.

9 . The method of claim 1 , wherein the video remix comprises a plurality of chunks each comprising a plurality of sub-sequences, preferably wherein at least one chunk is repeated in the video remix.

10 . The method of claim 1 , wherein the video remix comprises at least one chunk comprising sub-sequences from the first excerpt and at least one other chunk comprising sub-sequences from the second excerpt.

11 . The method of claim 1 , further including:

merging an audio track to the video remix, wherein the audio track corresponds to at least one of a specifically composed musical backing track matching the predetermined pattern and/or a musical backing track based on instrument tracks.

12 . The method of claim 1 , further including:

generating and adding a backing track from sounds from the input video.

13 . A data processing system comprising a processor configured to perform the method comprising:

receiving an input video;

selecting at least one excerpt from the input video, wherein an audio signal of the selected excerpt includes at least one onset, wherein the at least one excerpt has a predetermined length;

determining a plurality of sub-sequences of the at least one excerpt, wherein an n-th sub-sequence of the plurality of sub-sequences has a length corresponding to an n-th divisor of the predetermined length, wherein the n-th divisor equals to the (n−1)-th power of two, and n is an integer greater than two; and

rearranging the plurality of sub-sequences according to a predetermined pattern to form the video remix, wherein the video remix has a length corresponding to a multiple of the predetermined length.

14 . The data processing system of claim 13 , wherein the at least one excerpt begins at the at least one onset.

15 . The data processing system of claim 13 , wherein the method further includes:

detecting the at least one onset by applying an onset detection function to the input video.

16 . A non-transitory computer-readable medium comprising instruction which, when executed by a computer, cause the computer to carry out the method comprising:

receiving an input video;

selecting at least one excerpt from the input video, wherein an audio signal of the selected excerpt includes at least one onset, wherein the at least one excerpt has a predetermined length;

determining a plurality of sub-sequences of the at least one excerpt, wherein an n-th sub-sequence of the plurality of sub-sequences has a length corresponding to an n-th divisor of the predetermined length, wherein the n-th divisor equals to the (n−1)-th power of two, and n is an integer greater than two; and

rearranging the plurality of sub-sequences according to a predetermined pattern to form the video remix, wherein the video remix has a length corresponding to a multiple of the predetermined length.

17 . The non-transitory computer-readable medium of claim 16 , wherein the at least one excerpt begins at the at least one onset.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (17)