Techniques for selecting musical content for playback

Patent metadata

Field	Value
Publication number	US-9299331-B1
Application number	US-201314103444-A
Country	US
Kind code	B1
Filing date	Dec 11, 2013
Priority date	Dec 11, 2013
Publication date	Mar 29, 2016
Grant date	Mar 29, 2016

Abstract

Techniques are described for automatically selecting musical content for playback based on an initial “seed” of music selected by a user in a way that seamlessly extends the user's listening experience. The initially selected seed music might be, for example, an album or a playlist. Music that follows the seed music is algorithmically selected to match the music selected by the user.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving a plurality of first identifiers from a client device with one or more computing devices, each of the first identifiers corresponding to one of a plurality of first musical tracks of first musical content selected for playback on the client device, the first identifiers being received during playback of the first musical content on the client device;
using one or more of the computing devices, identifying a theme for the first musical content using first classification data associated with the first musical content, the first classification data being at least partially derived from first musical features extracted from each of the first musical tracks of the first musical content, the theme corresponding to a set of classification labels applied to a set of previously clustered musical tracks;
using one or more of the computing devices, identifying a plurality of second musical tracks of second musical content included in the set of previously clustered musical tracks corresponding to the theme for playback subsequent to playback of the first musical content, identification of the second musical content being achieved using second classification data associated with the second musical content, the second classification data being at least partially derived from second musical features extracted from each of the second musical tracks of the second musical content;
using one or more of the computing devices, retrieving the second musical content from a musical content data store;
using one or more of the computing devices, transmitting the second musical content to the client device; and
automatically initiating playback of the second musical content on the client device without human intervention upon completion of playback of the first musical content.

2. The method of claim 1 , wherein identifying the second musical content is further achieved using one or more of (1) behavior of a first user that selected the first musical content for playback in relation to the first or second musical content, (2) behavior of a population of users relating to the first or second musical content, or (3) metadata associated with the first or second musical content.

3. The method of claim 1 , further comprising identifying additional musical content corresponding to the theme for playback subsequent to playback of the second musical content, identification of the additional musical content being achieved using additional classification data associated with the additional musical content, the additional classification data being at least partially derived from additional musical features extracted from the additional musical content, the method further comprising repeating identification of the additional musical content until playback is interrupted by a user.

4. The method of claim 3 , further comprising receiving user feedback relating to playback of at least some of the second musical content or the additional musical content, wherein repeating identification of the additional musical content is done using the user feedback.

5. A system, comprising one or more computing devices configured to:
receive a first identifier representing first musical content selected for playback, the first identifier being received during playback of the first musical content;
identify a theme for the first musical content using first classification data associated with the first musical content, the first classification data being at least partially derived from first musical features extracted from the first musical content, the theme corresponding to a set of classification labels applied to a set of previously clustered musical tracks;
identify second musical content corresponding to the theme for playback subsequent to playback of the first musical content, identification of the second musical content being achieved using second classification data associated with the second musical content, the second classification data being at least partially derived from second musical features extracted from the second musical content; and
automatically initiate playback of the second musical content without human intervention upon completion of playback of the first musical content.

6. The system of claim 5 , wherein the one or more computing devices are further configured to identify the second musical content using one or more of (1) behavior of a first user that selected the first musical content for playback in relation to the first or second musical content, (2) behavior of a population of users relating to the first or second musical content, or (3) metadata associated with the first or second musical content.

7. The system of claim 5 , wherein the one or more computing devices are further configured to identify additional musical content corresponding to the theme for playback subsequent to playback of the second musical content, identification of the additional musical content being achieved using additional classification data associated with the additional musical content, the additional classification data being at least partially derived from additional musical features extracted from the additional musical content, and wherein the one or more computing devices are further configured to repeat identification of the additional musical content until playback is interrupted by a user.

8. The system of claim 7 , wherein the one or more computing devices are further configured to receive user feedback relating to playback of at least some of the second musical content or the additional musical content, and to repeat identification of the additional musical content using the user feedback.

9. The system of claim 8 , wherein the user feedback includes skipping a portion of the second musical content or the additional musical content, or rating the second musical content or the additional musical content.

10. The system of claim 5 , wherein the first musical content comprises one of a single musical track, a plurality of musical tracks, an album, or a playlist.

11. The system of claim 5 , wherein the first musical content and the second musical content are included in a catalog of musical content, and wherein the second musical content is only selected from a subset of most popular musical tracks in the catalog.

12. The system of claim 5 , wherein the second classification data are further derived using one or more of track metadata, user listening data, track consumption data, or track popularity data.

13. A client device, comprising one or more processors and memory configured to:
receive a first identifier representing first musical content selected for playback;
initiate playback of the first musical content;
during playback of the first musical content, transmit the first identifier for identification of a theme for the first musical content using first classification data associated with the first musical content, the first classification data being at least partially derived from first musical features extracted from the first musical content, the theme corresponding to a set of classification labels applied to a set of previously clustered musical tracks;
receive second musical content corresponding to the theme for playback subsequent to playback of the first musical content, identification of the second musical content being achieved using second classification data associated with the second musical content, the second classification data being at least partially derived from second musical features extracted from the second musical content; and
automatically initiate playback of the second musical content w

Assignees

• Amazon Tech Inc

Inventors

• Durham Brandon Scott
• Regmi Sarvesh Chandra
• Lazebny Vladimir
• Lutes Joshua Fredrick
• Malek Darren Levi
• Lu Haowei
• Phillips Christopher Woolf
• Brahmbhatt Kintan Dilipkumar
• Latin-Stoermer Toby Ray
• Ahern Gabriela
• Leslie William Jarred
• Di Xinjie
• Shepherd Benjamin Lee

Classifications

• G06F16/68

  Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually · CPC title

• H04N21/26258

  for generating a list of items to be played back in a given order, e.g. playlist, or scheduling item distribution according to such list (retrieval of multimedia data based on playlists G06F16/40) · CPC title

• G06F17/30772

  Physics · mapped topic

• G10H7/00 Primary

  Instruments in which the tones are synthesised from a data store, e.g. computer organs · CPC title

• G06F16/683 Primary

  using metadata automatically derived from the content · CPC title