Custom compilation videos
Described herein are, among other things, techniques, devices, and systems for generating one or more trained machine-learning models. Also described herein are techniques, devices, and systems for applying a consumption history of a particular user to the trained model(s) to generate score data indicating a correlation between each content-item title and the consumption history. The techniques then determine a ranked list of content items having a highest correlation to the consumption history, which may be used to retrieve videos associated with the most-correlated content items for generating a compilation video composed of these retrieved videos.
1. A method comprising:
determining, by a computing system, a history of one or more game titles played by a user;
inputting, by the computing system and into a trained machine-learning model, data indicative of the history;
generating, as output from the trained machine-learning model, score data indicating, for each of multiple available game titles, a score representing a correlation between the history and the respective available game title;
ranking the score data to generate a ranked list of the multiple available game titles;
determining, from the ranked list, an ordered set of available game titles having a highest correlation with the history;
retrieving, for each available game title from the ordered set, a trailer associated with the respective available game title from the ordered set; and
generating a compilation video using the respective trailers.
2. The method as recited in claim 1 , wherein the generating comprises generating the compilation video in an order defined by the ordered set such that a trailer of an available game title having a highest correlation with the history appears first in the compilation video.
3. The method as recited in claim 1 , further comprising identifying, from the ordered set, an available game title having a highest correlation with the history, and wherein the generating comprises generating the compilation video using audio associated with the available game title having the highest correlation with the history.
4. The method as recited in claim 1 , wherein the ordered set comprises a first ordered set and the compilation video comprises a first compilation video, and further comprising:
receiving feedback data indicating feedback from the user regarding the compilation video;
determining a second ordered set of available game titles based at least part on the score data and the feedback data;
retrieving, for each available game title from the second ordered set, a trailer associated with the respective available game title from the second ordered set; and
generating a second compilation video using the respective trailers associated with the respective available game titles from the second ordered set.
5. A method comprising:
determining, by a computing system, a history of one or more content items consumed by a user;
inputting, by the computing system and into a trained machine-learning model, data indicative of the history;
generating, as output from the trained machine-learning model, score data indicating, for each of multiple available content items, a score representing a correlation between the history and the respective available content item;
ranking the score data to generate a ranked list of the multiple available content items;
determining, from the ranked list, an ordered set of available content items having a highest correlation with the history;
retrieving, for each available content item from the ordered set, a video associated with the respective available content item from the ordered set; and
generating a compilation video using the respective videos.
6. The method as recited in claim 5 , wherein the generating comprises generating the compilation video in an order defined by the ordered set such that a video of an available content item having a highest correlation with the history appears first in the compilation video.
7. The method as recited in claim 5 , further comprising identifying, from the ordered set, an available content item having a highest correlation with the history, and wherein the generating comprises generating the compilation video using audio associated with the available content item having the highest correlation with the history.
8. The method as recited in claim 5 , wherein the ordered set comprises a first ordered set and the compilation video comprises a first compilation video, and further comprising:
receiving feedback data indicating feedback from the user regarding the compilation video;
determining a second ordered set of available content items based at least part on the score data and the feedback data;
retrieving, for each available content item from the second ordered set, a video associated with the respective available content item from the second ordered set; and
generating a second compilation video using the respective videos associated with the respective available content items from the second ordered set.
9. The method as recited in claim 5 , wherein the user comprises a first user, the score data comprises first score data, the ranked list comprises a first ranked list, the ordered set comprises a first ordered set, and further comprising:
determining, by the computing system, a history of one or more content items consumed by a second user;
inputting, by the computing system and into the trained machine-learning model, data indicative of the history of the one or more content items consumed by the second user;
generating, as output from the first trained machine-learning model, second score data indicating, for each of the multiple available content items, a score representing a correlation between the history of the one or more content items consumed by the second user and the respective available content item;
ranking the second score data to generate a second ranked list of the multiple available content items;
determining, from the second ranked list, a second ordered set of available content items having a highest correlation with the history of the one or more content items consumed by the second user;
retrieving, for each available content item from the second ordered set, a video associated with the respective available content item from the second ordered set; and
generating a compilation video using the respective videos associated with the respective available content item from the second ordered set.
10. The method as recited in claim 5 , further comprising presenting, with the compilation video, a control selectable by the user to acquire one or more content items of the ordered set.
11. The method as recited in claim 5 , wherein the one or more content items comprise one or more video games, one or more movies, one or more songs, or one or more electronic books.
12. The method as recited in claim 5 , wherein the retrieving comprises retrieving, for each available content item, a trailer associated with the respective available content item form the ordered set, and the generating comprises generating the compilation video using the respective trailers.
13. A computing system comprising:
one or more processors; and
one or more computer-readable media storing computer-executable instructions that, when executed, cause the one or more processors to perform acts comprising:
determining a history of one or more content items consumed by a user;
inputting, into a trained machine-learning model, data indicative of the history;
generating, as output from the trained machine-learning model, score data indicating, for each of multiple available content items, a score representing a correlation between the history and the respective available content item;
ranking the score data to generate a ranked list of the multiple available content items;
determining, from the ranked list, an ordered set of available content items having a highest correlation with the history;
retrieving, for each available content item from the ordered set, a video associated with the respective available content item from the ordered set; and
generating a compilation video using the respective videos.
14. The computing system as recited in claim 13 , wherein the generating comprises generating the compilation video in an order defined by the ordered set such that a video of an available content item having a highest correlation with the history appears first in the compilation video.
15. The computing system as recited in claim 13 , further comprising identifying, from the ordered set, an available content item having a highest correlation with the history, and wherein the generating comprises generating the compilation video using audio associated with the available content item having the highest correlation with the history.
16. The computing system as recited in claim 13 , wherein the ordered set comprises a first ordered set, the compilation video comprises a first compilation video, and the acts further comprise:
receiving feedback data indicating feedback from the user regarding the compilation video;
determining a second ordered set of available content items based at least part on the score data and the feedback data;
retrieving, for each available content item from the second ordered set, a video associated with the respective available content item from the second ordered set; and
generating a second compilation video using the respective videos associated with the respective available content items from the second ordered set.
17. The computing system as recited in claim 13 , wherein the user comprises a first user, the score data comprises first score data, the ranked list comprises a first ranked list, the ordered set comprises a first ordered set, and the acts further comprise:
determining a history of one or more content items consumed by a second user;
inputting, into the trained machine-learning model, data indicative of the history of the one or more content items consumed by the second user;
generating, as output from the first trained machine-learning model, second score data indicating, for each of the multiple available content items, a score representing a correlation between the history of the one or more content items consumed by the second user and the respective available content item;
ranking the second score data to generate a second ranked list of the multiple available content items;
determining, from the second ranked list, a second ordered set of available content items having a highest correlation with the history of the one or more content items consumed by the second user;
retrieving, for each available content item from the second ordered set, a video associated with the respective available content item from the second ordered set; and
generating a compilation video using the respective videos associated with the respective available content item from the second ordered set.
18. The computing system as recited in claim 13 , the acts further comprising presenting, with the compilation video, a control selectable by the user to acquire one or more content items of the ordered set.
19. The computing system as recited in claim 13 , wherein the one or more content items comprise one or more video games, one or more movies, one or more songs, or one or more electronic books.
20. The computing system as recited in claim 13 , wherein the retrieving comprises retrieving, for each available content item, a trailer associated with the respective available content item form the ordered set, and the generating comprises generating the compilation video using the respective trailers.