Communicating context to a device using an imperceptible audio identifier

Patent metadata

Field	Value
Publication number	US-2020082816-A1
Application number	US-201816123869-A
Country	US
Kind code	A1
Filing date	Sep 6, 2018
Priority date	Sep 6, 2018
Publication date	Mar 12, 2020
Grant date	—

Abstract

Various embodiments of systems and methods allow a system to embed an item identifier into a content item. A first device can then play an audio trigger that is imperceptible to humans before playing the item identifier. A second device can go into an active listening mode after detecting the audio trigger and record an audio segment contain the embedded item identifier. A system can then decode the item identifier to determine an appropriate context for the second device. The second device can then receive a vocal command or query and respond according to the determined context. In one example, the first device can be a television, and the second device can be a digital assistant (e.g., Amazon Alexa) that detects advertisements played on the television via audio signals embedded in accompanying audio streams. Subsequent user interactions with the digital assistant can then be informed by the context of the recently-heard advertisements.

First claim

1 . A computer-implemented method for using imperceptible context identifiers comprising:
receiving a first signal, at a service associated with a smart speaker, indicative of an inaudible audio trigger being presented in response to detection of a representation of an item in media content presented by a media presentation device; receiving a second signal, from the smart speaker, indicative of an audible audio segment presented by the media presentation device; determining that the audible audio segment corresponds to the item; receiving a third signal, from the smart speaker, indicative of an interaction request; determining a response to the interaction request, based at least upon the item represented in the media content; and transmitting instructions, to the smart speaker, to play the response.

2 . The computer-implemented method of claim 1 , wherein the audible audio segment has an ultrasonic component, the method further comprising:
receiving the ultrasonic component; and detecting an identifier for the item in the ultrasonic component.

3 . The computer-implemented method of claim 1 , further comprising:
generating the media content by adding the ultrasonic component to an intermediate media content; and transmitting, to the media presentation device, the audiovisual content.

4 . A computer-implemented method comprising:
receiving, at an audio processing server, a detected audio trigger from a source device, receiving, at the audio processing server and after receipt of the detected audio trigger, an audible audio segment from the source device; receiving, at the audio processing server, an interaction request uttered by a user; transmitting, to a media device from the audio processing server, information for an item determined to be associated with the audible audio segment; and generating a response to the interaction request based at least upon the item associated with the audible audio segment.

5 . The computer-implemented method of claim 4 , further comprising:
transmitting content for presentation with an audio component; determining a time that the item is featured in the content; presenting the audio trigger with the audio component at the time that the item is featured; and presenting the audible audio segment with the audio component after the audio trigger.

6 . The computer-implemented method of claim 5 , further comprising:
searching a database of items using a video frame of the content; and determining that the item associated with the audible audio segment is located within the video frame of the content.

7 . The computer-implemented method of claim 4 , further comprising:
transmitting an instruction to place the media device in a passive mode; and transmitting an instruction to place the media device in an active listening mode, after detecting the audio trigger.

8 . The computer-implemented method of claim 4 , further comprising:
determining that the interaction request is a request pertaining to a type of item; and determining that the item is of the type of item.

9 . The computer-implemented method of claim 4 , further comprising:
determining a location for the first source based on the audio trigger; and tuning a microphone system to isolate an audio stream from the detected location, wherein the audio stream comprises the audio segment.

10 . The computer-implemented method of claim 4 , further comprising:
determining a time that the audio trigger was detected; determining a time that the interaction request was detected; and determining that the difference between the time that the audio trigger was detected and the time that the interaction request was detected is less than a threshold amount.

11 . The computer-implemented method of claim 10 , further comprising:
receiving a request to associate the item with the audible audio segment, the request including the threshold amount.

12 . The computer-implemented method of claim 4 , further comprising:
determining a time in a video associated with the audible audio segment; and determining that the item is featured at the time in the video.

13 . The computer-implemented method of claim 4 , wherein the response includes at least one of an option to purchase the item, information related to one or more reviews, a summary rating of the item, a total time required to ship the item, or item details.

14 . A system, comprising:
at least one processor; and memory including instructions that, when executed by the at least one processor, cause the system to:
receive, at an audio processing server, a detected audio trigger from a source device,
receive, at the audio processing server and after receipt of the detected audio trigger, an audible audio segment from the source device;
receive, at the audio processing server, an interaction request from a second source device;
transmit, to the second source device from the audio processing server, information for an item determined to be associated with the audible audio segment; and
generate a response to the interaction request based at least upon the item associated with the audible audio segment.

15 . The system of claim 14 , wherein the instructions when executed further cause the system to:
transmit instructions to present content with an audio component; determine a time that the item is featured in the content; transmit instructions to present the audio trigger with the audio component at the time that the item is featured; and transmit instructions to present the audible audio segment with the audio component after the audio trigger.

16 . The system of claim 15 , wherein the instructions when executed further cause the system to:
search a database of items using a video frame of the content; and determine that an item of the database of items is located within the video frame of the content.

17 . The system of claim 14 , wherein the instructions when executed further cause the system to:
transmit an instruction to put a media device in a passive mode; and transmit an instruction to put the media device in an active listening mode after the media device detects the audio trigger.

18 . The system of claim 14 , wherein the instructions when executed further cause the system to:
determine that the interaction request is a request pertaining to a type of item; and determine that the item is of the type of item.

19 . The system of claim 14 , wherein the instructions when executed further cause the system to:
determine a time that the audio trigger was detected; determine a time that the interaction request was detected; and determine that the difference between the time that the audio trigger was detected and the time that the interaction request was detected is less than a threshold amount.

20 . The system of claim 14 , wherein the instructions when executed further cause the system to:
determine a time in a video associated with the audible audio segment; and determine that the item is featured at the time in the video.

Assignees

• Amazon Tech Inc

Inventors

• Mahajan Satayan

Classifications

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04N21/47

  End-user applications · CPC title

• G10L15/30

  Distributed recognition, e.g. in client-server systems, for mobile phones or network applications · CPC title

• G06F16/7834

  using audio features · CPC title

• G06F16/7867

  using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings · CPC title