Methods and systems for recommending content in context of a conversation

Patent metadata

Field	Value
Publication number	US-11140450-B2
Application number	US-201715824788-A
Country	US
Kind code	B2
Filing date	Nov 28, 2017
Priority date	Nov 28, 2017
Publication date	Oct 5, 2021
Grant date	Oct 5, 2021

Abstract

A media guidance application may monitor a conversation among users, and identify keywords in the conversation, without the use of wakewords. The keywords are used to search for media content that is relevant to the on-going conversation. Accordingly, the media guidance application presents relevant content to the users, during the conversation, to more actively engage the users. A conversation monitoring window may be used to present conversation information as well as relevant content. A listening mode may be used to manage when the media guidance application processes speech from a conversation. The media guidance application may access user profiles for keywords, select content types, select content sources, and determine relevancy of media content, to provide content in context of a conversation.

First claim

What is claimed is:

1. A method for recommending content in context of a conversation, the method comprising:
monitoring, with user equipment, verbal conversation between a plurality of users;
determining whether one of the plurality of users is logged into the user equipment;
in response to determining that one of the plurality of users is logged into the user equipment, activating a listening mode;
as the verbal conversation continues, processing words in the verbal conversation to selectively retrieve keywords from the words in the verbal conversation;
detecting the keywords in the verbal conversation without receiving a wakeword based on the selectively retrieved keywords;
searching a plurality of clips for a clip matching one of the keywords by:
comparing metadata associated with each of the plurality of clips with the detected keywords, and
selecting, based on the comparing, the clip of the plurality of clips having metadata matching a greater number of the keywords than another one of the plurality of clips having metadata matching at least one of the keywords; and

automatically, without receiving the wakeword, presenting the clip to the users during the verbal conversation.

2. The method of claim 1 further comprising:
identifying a content type associated with the detected keywords; and
selecting a content source to search based on the identified content type.

3. The method of claim 1 further comprising:
generating a verbal conversation processing window; and
temporarily storing words in the verbal conversation detected within the conversation processing window, such that words detected prior to a start time of the window are discarded.

4. The method of claim 1 , wherein the processing words in the verbal conversation is performed when the listening mode is active and is not performed when the listening mode is disabled.

5. The method of claim 1 , wherein each of the plurality of clips is less than a threshold duration.

6. The method of claim 1 , wherein the keywords relate to a segment in a media asset, and wherein each of the plurality of clips represents content in a scene of the media asset.

7. The method of claim 6 , wherein the segment is a scene or a portion of the media asset featuring a play in a sporting event.

8. The method of claim 1 , wherein automatically presenting the content to the users during the verbal conversation comprises changing content currently being presented on the user equipment to present the clip.

9. The method of claim 1 further comprising presenting some of the detected keywords with the clip with an indication that the clip was selected in response to detecting the keywords in the conversation.

10. The method of claim 1 , wherein selectively retrieving keywords comprises identifying words in the verbal conversation that match user profiles of each of the plurality of users.

11. A system for recommending content in context of a conversation, the system comprising:
a listening device configured to monitoring verbal conversation between a plurality of users; and
control circuitry configured to:
determine whether one of the plurality of users is logged into user equipment;
in response to determining that one of the plurality of users is logged into the user equipment, activate a listening mode;
as the verbal conversation continues, process words in the verbal conversation to selectively retrieve keywords from the words in the verbal conversation;
detect the keywords in the verbal conversation without receiving a wakeword based on the selectively retrieved keywords;
search a plurality of clips for a clip matching one of the keywords by:
comparing metadata associated with each of the plurality of clips with the detected keywords, and
selecting, based on the comparing, the clip of the plurality of clips having metadata matching a greater number of the keywords than another one of the plurality of clips having metadata matching at least one of the keywords; and

automatically, without receiving the wakeword, present the clip to the users during the verbal conversation.

12. The system of claim 11 , wherein the control circuitry is further configured to:
identify a content type associated with the detected keywords; and
select a content source to search based on the identified content type.

13. The system of claim 11 , wherein the control circuitry is further configured to:
generate a verbal conversation processing window; and
temporarily store words in the verbal conversation detected within the conversation processing window, such that words detected prior to a start time of the window are discarded.

14. The system of claim 11 , wherein the control circuitry is further configured to process words in the verbal conversation between the plurality of users when the listening mode is active and not process words in the verbal conversation between the plurality of users when the listening mode is disabled.

15. The system of claim 11 , wherein each of the plurality of clips is less than a threshold duration.

16. The system of claim 11 , wherein the keywords relate to a segment in a media asset, and wherein each of the plurality of clips represents content in a scene of the media asset.

17. The system of claim 16 , wherein the segment is a scene or a portion of the media asset featuring a play in a sporting event.

18. The system of claim 11 , wherein the control circuitry is further configured to change content currently being presented on the user equipment to present the clip.

19. The system of claim 11 , wherein the control circuitry is further configured to present some of the detected keywords with the clip with an indication that the clip was selected in response to detecting the keywords in the conversation.

20. The system of claim 11 , wherein the control circuitry is further configured to identify words in the verbal conversation that match user profiles of each of the plurality of users.

Assignees

• Rovi Guides Inc

Inventors

• Yelton Alexis
• Sharma Abhijeet
• Boonstra Peter

Classifications

• G10L2015/227

  of the speaker; Human-factor methodology · CPC title

• G10L15/1822

  Parsing for meaning understanding · CPC title

• G10L2015/225

  Feedback of the input speech · CPC title

• G06F16/683 Primary

  using metadata automatically derived from the content · CPC title

• H04N21/44213

  Monitoring of end-user related data (arrangements for monitoring the users' behaviour or opinions in broadcast systems H04H60/33) · CPC title