Systems and methods of generating personalized video clips for songs using a pool of short videos

1. A method for generating a personalized video for a song for a user, the method comprising:

identifying a plurality of video clips comprising audio of the song, wherein identifying the plurality of video clips further comprises:

retrieving metadata describing audio of available video clips;

determining, based on the metadata, particular video clips of the available video clips having metadata indicative of presence of audio of at least a portion of the song; and

identifying a subset of the particular video clips that are relevant to the user;

determining a portion of the song corresponding to each video clip of the plurality of video clips;

mapping each video clip of the plurality of video clips to a timeline of the song;

selecting, from the plurality of video clips, a subset of video clips comprising different portions of the song;

generating, from the subset of video clips, the personalized video for the song; and

generating the personalized video for display.

2. The method of claim 1 , wherein identifying the subset of the particular video clips that are relevant to the user further comprises:

accessing a popularity metric for each video clip of the particular video clips; and

identifying video clips of the particular video clips having a popularity metric above a threshold popularity.

3. The method of claim 1 , wherein identifying the subset of the particular video clips that are relevant to the user further comprises:

determining, for each video clip of the particular video clips, a content creator;

accessing a list of content creators followed by the user;

comparing the content creator of each video clip of the particular video clips with the list of content creators followed by the user; and

identifying, based on the comparing, video clips of the particular video clips created by content creators followed by the user.

4. The method of claim 1 , wherein identifying the subset of the particular video clips that are relevant to the user further comprises:

determining, for each video clip of the particular video clips, a content creator;

accessing a history of video clips liked by the user;

determining, based on the history of video clips liked by the user, a list of content creators whose video clips were liked by the user;

comparing the content creator of each video clip of the particular video clips with the list of content creators whose video clips were liked by the user; and

identifying, based on the comparing, video clips of the particular video clips created by content creators whose video clips were liked by the user.

5. The method of claim 1 , wherein identifying the subset of the particular video clips that are relevant to the user further comprises:

determining, for each video clip of the particular video clips, a content creator;

accessing a watch history of the user;

identifying, based on the watch history, a plurality of content creators the user may be interested in;

comparing the content creator of each video clip of the particular video clips with the plurality of content creators the user may be interested in; and

identifying, based on the comparing, video clips of the particular video clips created by content creators the user may be interested in.

6. The method of claim 1 , wherein identifying the subset of the particular video clips that are relevant to the user further comprises refraining from including, in the subset of the particular video clips, video clips created by a content creator previously blocked by the user.

7. The method of claim 1 , further comprising:

determining that a video clip has not been mapped to at least a portion of the song; and

in response to determining that a video clip as not been mapped to at least a portion of the song, selecting an advertisement for inclusion in the personalized video clip during the at least a portion of the song.

8. The method of claim 1 , wherein selecting, from the plurality of video clips, a subset of video clips comprising different portions of the song further comprises:

determining an entropy rating for each video clip of the plurality of video clips, wherein the entropy rating is based on an amount of movement in a video clip; and

selecting, from the plurality of video clips, a subset of video clips having similar entropy ratings.

9. The method of claim 1 , further comprising:

determining a first time signature of the song;

determining, for a video clip of the plurality of video clips, a second time signature of the audio of the song within the video clip;

calculating, based on a difference between the first time signature and the second time signature, a temporal scaling factor; and

modifying a duration of the video clip based on the temporal scaling factor.

10. A method for generating a personalized video for a song for a user, the method comprising:

identifying a plurality of video clips comprising audio of the song;

determining a portion of the song corresponding to each video clip of the plurality of video clips;

mapping each video clip of the plurality of video clips to a timeline of the song, wherein multiple video clips of the plurality of video clips are mapped to a first point in the timeline of the song, and wherein the audio of the song at the first point is repeated in the song;

selecting a subset of the multiple video clips that are mapped to the first point in the timeline of the song, a number of the subset of the multiple video clips corresponding to a number of times the audio of the song at the first point is repeated in the song;

mapping each video clip of the subset of the multiple video clips that are mapped to the first point in the timeline of the song to a different point in the timeline of the song at which the audio of the song at the first point is repeated;

selecting, from the plurality of video clips, a subset of video clips comprising different portions of the song;

generating, from the subset of video clips, the personalized video for the song; and

generating the personalized video for display.

11. The method of claim 10 , wherein the subset of video clips comprising different portions of the song, and the personalized video for the song, comprise the subset of the multiple video clips.

12. The method of claim 10 , wherein the subset of the multiple video clips comprise audio of a chorus of the song.

13. A system for generating a personalized video for a song for a user, the system comprising:

control circuitry configured to:

identify a plurality of video clips comprising audio of the song, wherein the control circuitry configured to identify the plurality of video clips is further configured to:

retrieve metadata describing the audio of available video clips;

determine, based on the metadata, a particular video clips of the available video clips having metadata indicate of presence of audio of at least a portion of the song; and

identify a subset of the particular video clips that are relevant to the user;

determine a portion of the song corresponding to each video clip of the plurality of video clips;

map each video clip of the plurality of video clips to a timeline of the song;

select, from the plurality of video clips, a subset of video clips comprising different portions of the song; and

generate, from the subset of video clips, the personalized video for the song; and

media output circuitry configured to generate the personalized video for display.

14. The system of claim 13 , wherein the control circuitry configured to identify the subset of the particular video clips that are relevant to the user is further configured to:

access a popularity metric for each video clip of the particular video clips; and

identify video clips of the particular video clips having a popularity metric above a threshold popularity.

15. The system of claim 13 , wherein the control circuitry configured to identify the subset of the particular video clips that are relevant to the user is further configured to:

determine, for each video clip of the particular video clips, a content creator;

access a list of content creators followed by the user;

compare the content creator of each video clip of the particular video clips with the list of content creators followed by the user; and

identify, based on the comparing, video clips of the particular video clips created by content creators followed by the user.

16. The system of claim 13 , wherein the control circuitry configured to identify the subset of the particular video clips that are relevant to the user is further configured to:

determine, for each video clip of the particular video clips, a content creator;

access a history of video clips liked by the user;

determine, based on the history of video clips liked by the user, a list of content creators whose video clips were liked by the user;

compare the content creator of each video clip of the particular video clips with the list of content creators whose video clips were liked by the user; and

identify, based on the comparing, video clips of the particular video clips created by content creators whose video clips were liked by the user.

17. The system of claim 13 , wherein the control circuitry configured to identify the subset of the particular video clips that are relevant to the user is further configured to:

determine, for each video clip of the particular video clips, a content creator;

access a watch history of the user;

identify, based on the watch history, a plurality of content creators the user may be interested in;

compare the content creator of each video clip of the particular video clips with the plurality of content creators the user may be interested in; and

identify, based on the comparing, video clips of the particular video clips created by content creators the user may be interested in.

18. The system of claim 13 , wherein the control circuitry configured to identify the subset of the particular video clips that are relevant to the user is further configured to refrain from including, in the subset of the particular video clips, video clips created by a content creator previously blocked by the user.

19. The system of claim 13 , wherein multiple video clips of the plurality of video clips are mapped to a first point in the timeline of the song and wherein audio of the song at the first point is repeated in the song, and wherein the control circuitry is further configured to:

select a subset of the multiple of video clips that are mapped to the first point in the timeline of the song, a number of the subset of the multiple video clips corresponding to a number of times the audio of the song at the first point is repeated in the song; and

map each video clip of the subset of the multiple video clips that are mapped to the first point in the timeline of the song to a different point in the timeline of the song at which the audio of the song at the first point is repeated.

20. The system of claim 13 , wherein the control circuitry is further configured to:

determine that a video clip has not been mapped to at least a portion of the song; and

in response to determining that a video clip as not been mapped to at least a portion of the song, select an advertisement for inclusion in the personalized video during the at least a portion of the song.

21. The system of claim 13 , wherein the control circuitry configured to select, from the plurality of video clips, a subset of video clips comprising different portions of the song is further configured to:

determine an entropy rating for each video clip of the plurality of video clips, wherein the entropy rating is based on an amount of movement in a video clip; and

select, from the plurality of video clips, a subset of video clips having similar entropy ratings.

22. The system of claim 13 , wherein the control circuitry is further configured to:

determine a first time signature of the song;

determine, for a video clip of the plurality of video clips a second time signature of the audio of the song within the video clip;

calculate, based on a difference between the first time signature and the second time signature, a temporal scaling factor; and

modify a duration of the video clip based on the temporal scaling factor.