IP Library Granted Patent US 10,019,996
Granted Patent B2
US 10,019,996 · App. 15/250,676 · Granted Jul 10, 2018

Orienting a microphone array to a user location

Inventors: Song Wang (Cary, NC); John Weldon Nicholson (Cary, NC); Ming Qian (Cary, NC)
Assignee: LENOVO (Singapore) PTE. LTD.
G10L17/22G10L17/04G10L17/18
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,019,996
App. No.
15/250,676
Granted
Jul 10, 2018
Kind
B2
Abstract

For orienting a microphone array to a user location, a processor detects a user location with a presence sensor that detects a user using electromagnetic signals. In addition, the processor orients a microphone array to the user location.

Claims (52)

1. An apparatus comprising:

a presence sensor that detects a user using electromagnetic signals;

a processor;

a memory that stores code executable by the processor to:

detect a user location with the presence sensor;

orient a microphone array to the user location;

receive a trigger from the user at the user location, wherein the trigger is one or more of an audible command, request, and prompt;

respond to the trigger;

determine the trigger is a false trigger in response to the trigger comprising a false trigger indication phrase from the user;

in response to determining the trigger is not a false trigger, interpret the trigger; and

in response to determining the trigger is a false trigger, record the false trigger to a learning structure.

2. The apparatus of claim 1 , wherein the presence sensor is one or more of a thermopile, a camera, and a network interface.

3. The apparatus of claim 1 , wherein the code is further executable by the processor to:

receive the trigger from the user at the user location of a plurality of user locations;

record a trigger timestamp and the user location for the trigger; and

respond to the trigger based on an order that the trigger was received.

4. The apparatus of claim 3 , wherein the code is further executable by the processor to respond to the trigger with a speaker oriented toward the user location.

5. The apparatus of claim 1 , wherein the learning structure comprises a plurality of subcomponents and a subcomponent weight for each subcomponent and the code is further executable by the processor to determine the trigger is false using an independent component analysis.

6. The apparatus of claim 1 , wherein the learning structure comprises a neural network, the false trigger is recorded using back propagation, and the code is further executable by the processor to determine the trigger is false from a neural network calculation.

7. The apparatus of claim 1 , wherein the learning structure comprises locations of other triggers and the code is further executable by the processor to determine the trigger is false based on subcomponents of a false trigger from a trigger location that is equivalent to a voice assistant location.

8. A method comprising:

detecting, by use of a processor, a user location with a presence sensor that detects a user using electromagnetic signals;

orienting a microphone array to the user location;

receiving a trigger from the user at the user location, wherein the trigger is one or more of an audible command, request, and prompt;

responding to the trigger;

determining the trigger is a false trigger in response to the trigger comprising a false trigger indication phrase from the user;

in response to determining the trigger is not a false trigger, interpreting the trigger; and

in response to determining the trigger is a false trigger, recording the false trigger to a learning structure.

9. The method of claim 8 , wherein the presence sensor is one or more of a thermopile, a camera, and a network interface.

10. The method of claim 8 , wherein the method further comprises:

receiving the trigger from the user at the user location of a plurality of user locations;

recording a trigger timestamp and the user location for the trigger; and

responding to the trigger based on an order that the trigger was received.

11. The method of claim 10 , wherein the method further comprises responding to the trigger with a speaker oriented toward the user location.

12. A program product comprising a non-transitory computer readable storage medium that stores code executable by a processor, the executable code comprising code to:

detect a user location with a presence sensor that detects a user using electromagnetic signals;

orient a microphone array to the user location;

receive a trigger from the user at the user location, wherein the trigger is one or more of an audible command, request, and prompt;

respond to the trigger;

determine the trigger is a false trigger in response to the trigger comprising a false trigger indication phrase from the user;

in response to determining the trigger is not a false trigger, interpret the trigger; and

in response to determining the trigger is a false trigger, record the false trigger to a learning structure.

13. The program product of claim 12 , wherein the presence sensor is one or more of a thermopile, a camera, and a network interface.

14. The program product of claim 12 , wherein the code further:

receives the trigger from the user at the user location of a plurality of user locations;

records a trigger timestamp and the user location for the trigger; and

responds to the trigger based on an order that the trigger was received.

15. The program product of claim 14 , wherein the code further responds to the trigger with a speaker oriented toward the user location.

16. The method of claim 8 , wherein the learning structure comprises a plurality of subcomponents and a subcomponent weight for each subcomponent and the method further determines the trigger is false using an independent component analysis.

17. The method of claim 8 , wherein the learning structure comprises a neural network, the false trigger is recorded using back propagation, and the method further determines the trigger is false from a neural network calculation.

18. The program product of claim 12 , wherein the learning structure comprises a plurality of subcomponents and a subcomponent weight for each subcomponent and the code further determines the trigger is false using an independent component analysis.

19. The program product of claim 12 , wherein the learning structure comprises a neural network, the false trigger is recorded using back propagation, and the code further determines the trigger is false from a neural network calculation.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 10, 2025
From: LENOVO PC INTERNATIONAL LIMITED
To: LENOVO SWITZERLAND INTERNATIONAL GMBH
Reel/Frame 069870/0670 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 8, 2019
From: LENOVO (SINGAPORE) PTE. LTD.
To: LENOVO PC INTERNATIONAL LIMITED
Reel/Frame 049690/0658 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 29, 2016
From: WANG, SONG; NICHOLSON, JOHN WELDON; QIAN, MING
To: LENOVO (SINGAPORE) PTE. LTD.
Reel/Frame 039569/0530 →
Continuity (1)
Related Publication 20180061422A1 · Mar 1, 2018
Cited By (2)
US 12,288,566 US 12,456,469