Automatically predicting relevant contexts for media items

Patent metadata

Field	Value
Publication number	US-12488044-B2
Application number	US-202117188858-A
Country	US
Kind code	B2
Filing date	Mar 1, 2021
Priority date	Jun 2, 2017
Publication date	Dec 2, 2025
Grant date	Dec 2, 2025

Abstract

The present technology pertains to automatically context labeling media items with relevant contexts, and further for algorithmically generating high quality playlists built around a context that are personalized to a profile of an account. This is accomplished by combining data from observed playlists, and data representing intrinsic properties of media items to predict contexts for media items.

First claim

The invention claimed is:

1 . At least one non-transitory computer readable medium comprising instructions that when executed cause one or more processors to:
determine an intrinsic representation for a first media item based on intrinsic properties of the first media item; identify one or more additional media items based on the one or more additional items being added to a same plurality of user-defined playlists as the first media item, wherein the plurality of user-defined playlist are defined by a plurality of users; determine behavioral characteristics for the first media item based on a frequency that the first media item is added to the same user-defined playlist as each of the one or more additional media items; determine a media item context for the first media it based on a combination representation for the first media item, wherein the combination representation is based on the intrinsic representation and the behavioral characteristics; and store, in a media item context association data store, an association between the context and the first media item, wherein the first media item is automatically selected for an automatically-created playlist based on the media item context.

2 . The at least one non-transitory computer readable medium of claim 1 , further comprising instructions that when executed cause the one or more processors to:
receive an input context from a client device associated with a user account of the user accounts, the input context originating from a request made by a user of the client device, wherein the client device has processed the request to determine the input context.

3 . The at least one non-transitory computer readable medium of claim 1 , wherein the instructions are effective to cause the one or more processors to:
provide the first media item in a candidate pool for the context; and create the automatically-created playlist for the context.

4 . The at least one non-transitory computer readable medium of claim 1 , wherein the instructions are effective to cause the one or more processors to:
create an intrinsic property representation for media items available from a media service, wherein the instructions to create the intrinsic property representation include instructions to cause the one or more processors to: analyze the media items available from the media service to determine physical properties of each media item; analyze the media items available from the media service to determine semantic characteristics of each media item; and combine the physical properties with the semantic characteristics to result in an intrinsic property representation.

5 . The at least one non-transitory computer readable medium of claim 4 , wherein the instructions to analyze the media items available from the media service to determine physical properties of each media item include instructions to:
analyze each media item for timbre properties; analyze each media item for rhythm properties; and output the physical properties representing the timbre properties and the rhythm properties.

6 . The at least one non-transitory computer readable medium of claim 1 , further comprising instructions to:
receive a text or speech input requesting a playlist targeted to an input context, wherein the media items are selecting based on having a media-item-context association with the input context greater than a threshold in accordance with the combination representation.

7 . The at least one non-transitory computer readable medium of claim 1 , further comprising instructions to:
include, in the combination representation, media items not associated with any context in the combination representation based on a predicted context.

8 . A system comprising:
one or more processors; and at least one non-transitory computer readable medium comprising instructions that when executed by the one or more processors cause a computing system to:
determine an intrinsic representation for a first media item based on intrinsic properties of the first media item;
identify one or more additional media items based on the one or more additional items being added to a same plurality of user-defined playlists as the first media item, wherein the plurality of user-defined playlist are defined by a plurality of users;
determine behavioral characteristics for the first media item based on a frequency that the first media item is added to the same user-defined playlist as each of the one or more additional media items;
determine a media item context for the first media it based on a combination representation for the first media item, wherein the combination representation is based on the intrinsic representation and the behavioral characteristics; and
store, in a media item context association data store, an association between the context and the first media item,
wherein the first media item is automatically selected for an automatically-created playlist based on the media item context.

9 . The system of claim 8 , wherein the instructions are effective to cause the computing system to:
provide the first media item in a candidate pool for the context.

10 . The system of claim 9 , wherein the instructions are effective to cause the computing system to:
determine media items from the candidate pool for the context that are compatible with a media item profile of a user account of the user accounts, for which a playlist is being created; and create the playlist from the media items that are compatible with the media item profile of the account based on the context.

11 . The system of claim 8 , wherein the instructions are effective to cause the computing system to:
create an intrinsic property representation for media items available from a media service, wherein the instructions to create the intrinsic property representation include instructions to cause the computing system to: analyze the media items available from the media service to determine physical properties of each media item; analyze the media items available from the media service to determine semantic characteristics of each media item; and combine the physical properties with the semantic characteristics to result in an intrinsic property representation.

12 . The system of claim 11 , wherein the instructions to analyze the media items available from the media service to determine physical properties of each media item include instructions to:
analyze each media item for timbre properties; analyze each media item for rhythm properties; and output the physical properties representing the timbre properties and the rhythm properties.

13 . The system of claim 8 , further comprising instructions to:
receive a text or speech input requesting a playlist targeted to an input context, wherein the media items are selecting based on having a media-item-context association with the input context greater than a threshold in accordance with the combination representation.

14 . The system of claim 8 , further comprising instructions to:
include, in the combination representation, media items not associated with any context in the combination representation based on a predicted context.

15 . A method, comprising:
determining an intrinsic representation for a first media item based on intrinsic properties of the first media item; identifying one or more additional media items based on the one or more additional items being added to a same plurality of user-defined playlists as the first media item, wherein the plurality of user-defined playlist are defined by a plurality of users; determ

Assignees

• Apple Inc

Inventors

• Levy Mark H
• Shenoy Arvind S
• Mauch Matthias
• Cartoon Daniel

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N20/20

  Ensemble learning · CPC title

• G06F16/639

  using playlists · CPC title

• G06F16/635

  Filtering based on additional data, e.g. user or group profiles · CPC title

• G06N3/084 Primary

  Backpropagation, e.g. using gradient descent · CPC title