Multiple stream tuning

Patent metadata

Field	Value
Publication number	US-9936239-B2
Application number	US-201615195203-A
Country	US
Kind code	B2
Filing date	Jun 28, 2016
Priority date	Jun 28, 2016
Publication date	Apr 3, 2018
Grant date	Apr 3, 2018

Abstract

System and techniques for multiple stream tuning are described herein. A plurality of content streams may be received. The plurality of content streams may be ranked. A user attention level may be scored as the user observes at least one content stream of the plurality of content streams. The user attention level and a rank for the at least one content stream may be compared to remaining ranks of the plurality of content streams to produce a difference. A stream action may be performed on a set of content streams from the plurality of content streams based on the difference.

First claim

What is claimed is:

1. system for multiple stream tuning, the system comprising:
a decoder to receive a plurality of content streams;
a content sensor array to rank the plurality of content streams, wherein to rank the plurality of content streams includes, for each of the plurality of content streams, the content sensor is to apply a set of assays to a content stream to produce a set of artifacts;
a sensor to score a user attention level as the user observes at least one content stream of the plurality of content streams;
a comparator to compare the user attention level and a rank for the at least one content stream to remaining ranks of the plurality of content streams to produce a difference; and
a switch to:
perform a stream action on a set of content streams from the plurality of content streams based on the difference;
evaluate the set of artifacts in a user context for the user to produce artifact evaluations, wherein the user context includes a matching group to a set of users, and wherein to evaluate the set of artifacts in a user context includes the switch to use a portion of a second user context from a member of the set of users to produce the artifact evaluations; and
modify a rank for the content stream based on the artifact evaluations.

2. The system of claim 1 , wherein a member of set of assays is automatic speech recognition (ASR), and wherein a member of the set of artifacts that correspond to the ASR assay is a string.

3. The system of claim 1 , wherein a member of the set of assays is a visual classifier, and wherein a member of the set of artifacts include an object identification corresponding to the output of the visual classifier.

4. The system of claim 2 , wherein to evaluate the set of artifacts in the user context includes the switch to perform a semantic comparison between text in the string and tokens in the user context, the tokens including a score corresponding to user action in regard to subject matter corresponding to the tokens.

5. The system of claim 4 , wherein the sensor is to:
observe the user action in a sample period;
calculate an attention score for the user action; and
wherein the switch is to:
correlate the attention score to one or more tokens from a content stream during the sample period; and
update the user context to include a modified attention score for the one or more tokens.

6. The system of claim 1 , wherein, given a second content stream in the plurality of content streams, the stream action is at least one of the switch to:
replace a current content stream on a screen with the second content stream when the second content stream has a higher rank than the current content stream;
replace a current audio stream with an audio stream from a second content stream that has a higher rank that the current audio stream;
place the second content stream in a picture-in-picture container on a screen when the second content;
record the second content stream;
modify the volume of an audio portion of the second content stream; or
render a notification to the user regarding the second content stream.

7. A method for multiple stream tuning, the method comprising:
receiving a plurality of content streams;
ranking the plurality of content streams, wherein ranking the plurality of content streams includes, for each of the plurality of content streams:
applying a set of assays to a content stream to produce a set of artifacts;
evaluating the set of artifacts in a user context for the user to produce artifact evaluations, wherein the user context includes a matching group to a set of users, and wherein evaluating the set of artifacts in a user context includes using a portion of a second user context from a member of the set of users to produce the artifact evaluations; and
modifying a rank for the content stream based on the artifact evaluations;

scoring a user attention level as the user observes at least one content stream of the plurality of content streams;
comparing the user attention level and a rank for the at least one content stream to remaining ranks of the plurality of content streams to produce a difference; and
performing a stream action on a set of content streams from the plurality of content streams based on the difference.

8. The method of claim 7 , wherein a member of set of assays is automatic speech recognition (ASR), and wherein a member of the set of artifacts that correspond to the ASR assay is a string.

9. The method of claim 7 , wherein a member of the set of assays is a visual classifier, and wherein a member of the set of artifacts include an object identification corresponding to the output of the visual classifier.

10. The method of claim 8 , wherein evaluating the set of artifacts in the user context includes performing a semantic comparison between text in the string and tokens in the user context, the tokens including a score corresponding to user action in regard to subject matter corresponding to the tokens.

11. The method of claim 10 , comprising:
observing the user action in a sample period;
calculating an attention score for the user action;
correlating the attention score to one or more tokens from a content stream during the sample period; and
updating the user context to include a modified attention score for the one or more tokens.

12. The method of claim 7 , wherein, given a second content stream in the plurality of content streams, the stream action is at least one of:
replacing a current content stream on a screen with the second content stream when the second content stream has a higher rank than the current content stream;
replacing a current audio stream with an audio stream from a second content stream that has a higher rank that the current audio stream;
placing the second content stream in a picture-in-picture container on a screen when the second content;
recording the second content stream;
modifying the volume of an audio portion of the second content stream; or
rendering a notification to the user regarding the second content stream.

13. At least one non-transitory machine readable medium including instructions for multiple stream tuning, the instructions, when executed by a machine, causing the machine to:
receive a plurality of content streams;
rank the plurality of content streams, wherein, to rank the plurality of content streams includes, for each of the plurality of content streams, includes the instructions to cause the machine to:
apply a set of assays to a content stream to produce a set of artifacts;
evaluate the set of artifacts in a user context the user to produce artifact evaluations, wherein the user context includes a matching group to a set of users, and wherein, evaluate the set of artifacts in a user context, the machine is to us a portion of a second user context from a member of the set of users to produce the artifact evaluations; and
modify a rank for the content stream based on the artifact evaluations;

score a user attention level as the user observes at least one content stream of the plurality of content streams;
compare the user attention level and a rank for the at least one content stream to remaining ranks of the plurality of content streams to produce a difference; and
perform a stream action on a set of content streams from the plurality of content streams based on the difference.

14. The at least one machine readable medium of claim 13 , wherein a member of set of assays is automatic speech recognition (ASR), and wherein a member of the set of artifacts that correspond to the ASR assay is a string.

15. The at least one machine r

Assignees

• Intel Corp

Inventors

• Pogorelik Oleg
• Wouhaybi Rita H
• Nayshtut Alex

Classifications

• G06F16/683

  using metadata automatically derived from the content · CPC title

• H04N21/44008

  involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream (arrangements characterised by components specially adapted for monitoring, identification or recognition of video in broadcast systems H04H60/59) · CPC title

• H04N21/4784

  receiving rewards · CPC title

• H04N21/4532

  involving end-user characteristics, e.g. viewer profile, preferences (monitoring of user activities for profile generation for accessing a video database G06F16/739; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18) · CPC title

• H04N21/4756

  for rating content, e.g. scoring a recommended movie · CPC title