Identifying media content

• US 10,055,413 B2
• Filed: 11/18/2015
• Issued: 08/21/2018
• Est. Priority Date: 05/19/2015
• Status: Active Grant

• Current Assignee: Spotify AB (Spotify Technology SA)
• Original Assignee: Spotify AB (Spotify Technology SA)
• Inventors: Jehan, Tristan, Dziuk, Dariusz, Soderstrom, Gustav, Rando, Mateo, Montecchio, Nicola
• Primary Examiner(s): Tran, Loc
• Application Number:
  US14/945,008
• Publication Number:
  US 20160342598A1
• Time in Patent Office: 1,007 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/4387 by the use of playlists
• CPC Class Codes: G06F 16/636 by using biological or physiological data
• CPC Class Codes: G06F 16/639 using playlists
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: G11B 27/031 Electronic editing of digitised analogue information signals, e.g. audio or video signals
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence
• CPC Class Codes: G11B 27/329 on a disc [VTOC]

Abstract

Systems, devices, apparatuses, components, methods, and techniques for identifying media content for playback during a repetitive motion activity are provided. An example media-playback device includes a media-output device that plays media content items and a repetitive-motion activity content identification engine. The repetitive-motion activity content identification engine is configured to: review media content items to identify the media content items that are conducive to performing repetitive-motion activities; and select certain media content items from the media content items, the certain media content items being conducive to performing the repetitive-motion activities.

Claims

1. A media system comprising:

at least one hardware processing device;

a media-output device that plays media content items; and

at least one computer readable data storage device storing data instructions that, when executed by the at least one hardware processing device, cause the media system to;

identify, from a plurality of playlists, a subset of playlists containing one or more repetitive-motion activity related keywords appearing in descriptions associated with the subset of playlists, wherein the plurality of playlists include media content items;

evaluate a plurality of media content items included in the subset of playlists, and identify based on a playlist inclusion score a plurality of positive example media content items from the plurality of media content items for playback during repetitive motion activities, wherein the playlist inclusion score is calculated as a difference between a number of playlists that include the media content item, from the plurality of playlists, that are conducive to performing the repetitive-motion activities and a number of playlists that include the media content item, from the plurality of playlists, that are inappropriate for performing the repetitive-motion activities;

generate a model usable to characterize a suitability of media content items for repetitive motion activities based on characteristics of audio signals of the identified plurality of positive example media content items that have playlist inclusion scores that exceed a threshold;

review media content items to identify the media content items that are conducive to performing repetitive-motion activities; and

select certain media content items from the media content items based upon comparison to the generated model, the certain media content items being conducive to performing the repetitive-motion activities.

2. The media system of claim 1, comprising further data instructions that cause the processing device to:

identify potential example media content items, the potential example media content items being identified based on inclusion in a playlist having characteristics that are conducive to performing the repetitive-motion activities; and

calculate a combined playlist inclusion score for at least one of the potential example media content items based on characteristics of playlists in which the potential example media content item is included.

3. The media system of claim 2, comprising further data instructions that cause the processing device to:

review an aspect associated with a plurality of playlists to identify the playlist.

4. The media system of claim 3, wherein the aspect is one or more words in a title of the playlist.

5. The media system of claim 1, comprising further data instructions that cause the processing device to:

calculate a score for each of the media content items; and

select certain media content items based upon the score.

6. The media system of claim 1, comprising further data instructions that cause the processing device to:

use a repetitive motion activity model to analyze audio signals associated with each of the media content items; and

compare aspects of the audio signals to the repetitive-motion activity model to determine if each of the media content items is conducive to performing the repetitive-motion activities.

7. The media system of claim 6, wherein the aspects of the audio signals include one or more of an average musical event duration, a tempo regularity, a percussivity, and a beat strength.

8. A method for selecting media content that is suitable for running, the method comprising:

identifying, from a plurality of playlists, a subset of playlists containing one or more running related keywords appearing in descriptions associated with the subset of playlists, wherein the plurality of playlists include media content items;

evaluating a plurality of media content items included in the subset of playlists, and identifying based on a playlist inclusion score a plurality of positive example media content items from the plurality of media content items for playback during running, wherein the playlist inclusion score is calculated as a difference between a number of playlists that include the media content item, from the plurality of playlists, that are conducive to running and a number of playlists that include the media content item, from the plurality of playlists, that are inappropriate for running;

generating a model usable to characterize a suitability of media content items for running based on characteristics of audio signals of the identified plurality of positive example media content items that have playlist inclusion scores that exceed a threshold;

reviewing a list of media content items;

comparing each of the media content items in the list of media content items to the generated model;

identifying those media content items that are suitable for running based upon comparison to the generated model; and

presenting those media content items to a user.

9. The method of claim 8, further comprising:

identifying a playlist of potential example media content items, the playlist having characteristics that are suitable for running;

reviewing the potential example media content items to identify media content items that are not suitable for running;

selecting certain media content items from the potential example media content items, the certain media content items being suitable for running; and

developing the model based upon the certain media content items.

10. The method of claim 9, further comprising using machine learning techniques to build the model.

11. The method of claim 9, further comprising calculating a combined playlist inclusion score for at least one of the potential example media content items based on characteristics of playlists in which the potential example media content item is included.

12. The method of claim 9, wherein identifying the playlist further comprises reviewing an aspect associated with a plurality of playlists to identify the playlist.

13. The method of claim 12, wherein the aspect is a title of the playlist.

14. The method of claim 8, further comprising:

calculating a score for each of the media content items using the model; and

selecting certain media content items based upon the score.

15. The method of claim 14, further comprising analyzing audio signals associated with each of the media content items to determine the score for each of the media content items in the list.

16. The method of claim 15, wherein analyzing the audio signals includes examining one or more of an average musical event duration, a tempo regularity, a percussivity, and a beat strength.

17. A method for selecting media content that is suitable for running, the method comprising:

identifying, from a plurality of playlists, a subset of playlists comprising a plurality of potential example media content items based on one or more running-related keywords appearing in descriptions associated with the subset of playlists, wherein the subset of playlists have characteristics that are suitable for running;

evaluating the subset of playlists, and selecting based on a playlist inclusion score positive example media content items from the potential example media content items, the positive example media content items being suitable for running, the playlist inclusion score is calculated as a difference between a number of playlists that include the potential example media content items, from the plurality of playlists, that are conducive to running and a number of playlists that include the potential example media content items, from the plurality of playlists, that are inappropriate for running;

developing a model usable to characterize a suitability of media content items for running based upon characteristics of audio signals of the identified plurality of positive example media content items that have playlist inclusion scores that exceed a threshold;

reviewing a list of media content items;

comparing each of the media content items in the list of media content items to the model;

identifying those media content items in the list of media content items that are suitable for running based upon comparison to the model; and

presenting those media content items to a user.

18. The method of claim 17, further comprising using machine learning techniques to build the model.

19. The method of claim 17, further comprising calculating a combined playlist inclusion score for at least one of the potential example media content items based on characteristics of playlists in which the potential example media content item is included.

20. The method of claim 17, further comprising:

calculating a score for each of the media content items in the list; and

analyzing audio signals associated with each of the media content items to determine the score for each of the media content items in the list, wherein analyzing the audio signals includes examining one or more of an average musical event duration, a tempo regularity, a percussivity, and a beat strength.