Granted Patent B2

US 12,266,360 · App. 18/500,836 · Granted Apr 1, 2025

Inventors: Edison Lin (Los Altos Hills, CA); Rowena Young (Menlo Park, CA); Lauren Palmateer (San Francisco, CA)

Assignee: Adeia Guides Inc.

G10L15/22G10L21/0316H04R1/326G10L2015/223

Abstract

Claims (56)

1. A method comprising:

receiving, at a user device, a signature sound sequence, wherein the signature sound sequence has a frequency outside audible frequency for humans;

determining a threshold metric based on the signature sound sequence;

receiving, at the user device, a voice command;

determining a metric for the voice command;

determining, based on comparing the metric for the voice command to the threshold metric based on the signature sound sequence, that the metric for the voice command is different than the threshold metric; and

causing the voice command to be executed.

2. The method of claim 1 , wherein the determining the threshold metric based on the signature sound sequence further comprises:

determining a first gain of the signature sound sequence at a first time;

determining a second gain of the signature sound sequence at a second time;

determining a reference gain by averaging the first gain and the second gain;

normalizing the first gain based on the reference gain and normalizing the second gain based on the reference gain; and

determining the threshold metric based on a difference between the normalized first gain and the normalized second gain.

3. The method of claim 1 , wherein the signature sound sequence is received via a unidirectional microphone of the user device.

4. The method of claim 1 , wherein the voice command is received via a unidirectional microphone of the user device.

5. The method of claim 1 , wherein the signature sound sequence is transmitted by a speaker of the user device.

6. The method of claim 1 , further comprising:

determining, based on comparing the metric for the voice command to the threshold metric, that the metric for the voice command is close in value to the threshold metric;

determining that the voice command is a false voice command; and

ignoring the false voice command.

7. The method of claim 6 , wherein the false voice command indicates that the voice command is generated from a speaker at a current location of the user device.

8. The method of claim 1 , wherein the signature sound sequence and the voice command are received at a first time during a setup procedure for the user device.

9. The method of claim 1 , wherein causing the voice command to be executed comprises:

determining a domain of the voice command;

determining a target device for the voice command based on the domain; and

executing the voice command with respect to the target device.

10. A system comprising:

input/output circuitry configured to:

receive, at a user device, a signature sound sequence, wherein the signature sound sequence has a frequency outside audible frequency for humans;

processing circuitry configured to:

determine a threshold metric based on the signature sound sequence;

wherein the input/output circuitry is further configured to:

receive, at the user device, a voice command;

wherein the processing circuitry is further configured to:

determine a metric for the voice command;

determine, based on comparing the metric for the voice command to the threshold metric based on the signature sound sequence, that the metric for the voice command is different than the threshold metric; and

cause the voice command to be executed.

11. The system of claim 10 , wherein the processing circuitry is further configured to determine the threshold metric based on the signature sound sequence by:

determining a first gain of the signature sound sequence at a first time;

determining a second gain of the signature sound sequence at a second time;

determining a reference gain by averaging the first gain and the second gain;

normalizing the first gain based on the reference gain and normalizing the second gain based on the reference gain; and

determining the threshold metric based on a difference between the normalized first gain and the normalized second gain.

12. The system of claim 10 , wherein the input/output circuitry is further configured to receive the signature sound sequence via a unidirectional microphone of the user device.

13. The system of claim 10 , wherein the input/output circuitry is further configured to receive the voice command via a unidirectional microphone of the user device.

14. The system of claim 10 , wherein the processing circuitry is further configured to transmit the signature sound sequence by a speaker of the user device.

15. The system of claim 10 , wherein the processing circuitry is further configured to:

determine, based on comparing the metric for the voice command to the threshold metric, that the metric for the voice command is close in value to the threshold metric;

determine that the voice command is a false voice command; and

ignore the false voice command.

16. The system of claim 15 , wherein the false voice command indicates that the voice command is generated from a speaker at a current location of the user device.

17. The system of claim 10 , wherein the input/output circuitry is further configured to receive the signature sound sequence and the voice command at a first time during a setup procedure for the user device.

18. The system of claim 10 , wherein causing, by the processing circuitry, the voice command to be executed comprises:

determining a domain of the voice command;

determining a target device for the voice command based on the domain; and

executing the voice command with respect to the target device.

Assignments (2)

Continuity (3)

References Cited (21)