INTELLIGENT AUTOMATED ASSISTANT FOR MEDIA EXPLORATION

• US 20170358302A1
• Filed: 09/15/2016
• Published: 12/14/2017
• Est. Priority Date: 06/08/2016
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: ORR, Ryan M., BERNARDO, Matthew P., MANDEL, Daniel J.
• Granted Patent: US 10,049,663 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/433 using audio data
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 25/84 for discriminating voice from noise

Abstract

Systems and processes for operating an intelligent automated assistant to explore media items are provided. In one example process, a speech input representing a request for one or more media items is received from a user. The process determines whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items. In response to determining that the speech input corresponds to a user intent of obtaining personalized recommendations for media items, at least one media item is obtained from a user-specific corpus of media items. The user-specific corpus of media items is generate based on data associated with the user. The at least one media item is provided.

Claims

1. A system for operating a digital assistant to explore media items, the system comprising:

one or more processors; and

memory having instructions stored thereon, the instructions, when executed by the one or more processors, cause the one or more processors to;

receive, from a user, speech input representing a request for one or more media items;

determine whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items; and

in response to determining that the speech input corresponds to a user intent of obtaining personalized recommendations for media items;

obtain at least one media item from a user-specific corpus of media items, the user-specific corpus of media items generated based on data associated with the user; and

provide the at least one media item.

2. The system of claim 1, wherein determining whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items comprises determining whether a number of parameters defined in the speech input is less than a threshold number.

3. The system of claim 1, wherein determining whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items comprises determining whether the speech input includes one of a plurality of phrases corresponding to the user intent of obtaining personalized recommendations for media items.

4. The system of claim 1, wherein determining whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items comprises determining whether the speech input refers to the user.

5. The system of claim 1, wherein the instructions further cause the one or more processors to:

rank a plurality of candidate media items from the user-specific corpus of media items using a user-specific media ranking model, the user-specific media ranking model generated based on a plurality of previous media-related requests from the user, wherein obtaining the at least one media item includes selecting the at least one media item from the plurality of candidate media items based on the ranking.

6. The system of claim 5, wherein the instructions further cause the one or more processors to:

receive, from the user, a second speech input;

determine whether the second speech input corresponds to a rejection of the at least one media item; and

in response to determining that the second speech input corresponds to a rejection of the at least one media item;

update the user-specific media ranking model in accordance with the rejection.

7. The system of claim 6, wherein the instructions further cause the one or more processors to:

re-rank the plurality of candidate media items from the user-specific corpus of media items based on the rejection of the at least one media item; and

select at least one second media item from the plurality of candidate media items based on the re-ranking.

8. The system of claim 1, wherein each media item in the user-specific corpus of media items includes metadata that indicates an activity associated with the media item, and wherein the activity is associated with the media item based on a music tempo of the media item.

9. The system of claim 1, wherein each media item in the user-specific corpus of media items includes metadata that indicates a mood associated with the media item, and wherein the mood is associated with the media item based on a music key of the media item.

10. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine whether the speech input defines an occasion associated with a time period; and

in response to determining that the speech input defines an occasion associated with a time period, obtain the at least one media item based the occasion, wherein the at least one media item includes metadata indicating the occasion.

11. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine whether the speech input defines an editorial list associated with a media establishment; and

in response to determining that the speech input defines an editorial list associated with a media establishment, obtain the at least one media item based on the editorial list associated with the media establishment, wherein the at least one media item includes metadata indicating the editorial list associated with the media establishment.

12. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine whether the speech input defines a mood; and

in response to determining that the speech input defines a mood, obtain the at least one media item based on the mood, wherein the at least one media item includes metadata indicating the mood.

13. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine whether the speech input defines an activity; and

in response to determining that the speech input defines an activity, obtain the at least one media item based on the activity, wherein the at least one media item includes metadata indicating the activity.

14. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine whether the speech input defines a time period;

in response to determining that the speech input defines a time period, determine whether the speech input defines a genre in association with the time period; and

in response to determining that the speech input defines a genre in association with the time period, determine a subgenre based on the time period and the genre, wherein the at least one media item is obtained based on the subgenre, and wherein the at least one media item includes metadata indicating the subgenre.

15. The system of claim 1, wherein the speech input defines a category of media items, wherein obtaining the at least one media item includes obtaining a plurality of media items associated with the category of media items, and wherein the instructions further cause the one or more processors to:

determine a degree of familiarity of the user with the category of media items, wherein an average popularity rating of the plurality of media items is based on the degree of familiarity of the user with the category of media items.

16. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine an identity of the user by performing speaker identification using the speech input; and

based on the determined identity of the user, determine the user-specific corpus of media items from a plurality of user-specific corpuses of media items.

17. The system of claim 1, wherein obtaining the at least one media item comprises sending an encrypted token to a remote server, the encrypted token containing user identification information, and wherein the encrypted token is required to access the user-specific corpus of media items via the remote server.

18. The system of claim 1, wherein the instructions further cause the one or more processors to:

determine whether a domain corresponding to the speech input is one of a plurality of predetermined domains;

in response to determining that a domain corresponding to the speech input is one of a plurality of predetermined domains;

upon providing the at least one media item, receive audio input;

determine whether the audio input contains speech; and

in response to determining that the audio input does not contain speech, cease to receive audio input after a predetermined amount of time.

19. The system of claim 18, wherein the instructions further cause the one or more processors to:

in response to determining that the audio input contains speech;

determine whether the speech of the audio input corresponds to a same domain as the speech input;

in response to determining that the speech of the audio input corresponds to a same domain as the speech input;

determine a user intent corresponding to the speech of the audio input; and

provide a response to the audio input in accordance with the user intent corresponding to the speech of the audio input.

20. The system of claim 19, wherein the instructions further cause the one or more processors to:

in response to determining that the speech of the audio input does not correspond to a same domain as the speech input;

forgo providing a response to the audio input.

21. The system of claim 18, wherein the predetermined amount of time is based on a degree of ambient noise detected in the audio input.

22. The system of claim 1, wherein providing the at least one media item comprises playing a media item of the at least one media item, and wherein the instructions further cause the one or more processors to:

while playing the media item, receive a third speech input;

determine, based on the playing media item and the third speech input, a user intent corresponding to the third speech input; and

provide a response in accordance with the user intent that corresponds to the third speech input.

23. The system of claim 1, wherein the instructions further cause the one or more processors to:

in response to determining that the speech input does not correspond to a user intent of obtaining personalized recommendations for media items;

determine whether the speech input corresponds to a user intent of obtaining media items having a recent release date; and

in response to determining that the speech input corresponds to a user intent of obtaining media items having a recent release date;

obtain at least one second media item from a second corpus of media items, wherein each media item in the second corpus of media items has a release date that is within a predetermined time range of a current date; and

provide the at least one second media item.

24. A method for operating a digital assistant to explore media items, the method comprising:

at an electronic device with one or more processors and memory;

receiving, from a user, speech input representing a request for one or more media items;

determining whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items; and

in response to determining that the speech input corresponds to a user intent of obtaining personalized recommendations for media items;

obtaining at least one media item from a user-specific corpus of media items, the user-specific corpus of media items generated based on data associated with the user; and

providing the at least one media item.

25. A non-transitory computer-readable medium having instructions stored thereon, the instructions, when executed by one or more processors, cause the one or more processors to:

receive, from a user, speech input representing a request for one or more media items;

determine whether the speech input corresponds to a user intent of obtaining personalized recommendations for media items; and

in response to determining that the speech input corresponds to a user intent of obtaining personalized recommendations for media items;

obtain at least one media item from a user-specific corpus of media items, the user-specific corpus of media items generated based on data associated with the user; and

provide the at least one media item.