Granted Patent B1

US 10,536,288 · App. 15/841,016 · Granted Jan 14, 2020

Inventors: Jonathan Alan Leblang (Menlo Park, CA); Milo Oostergo (Issaquah, WA); James L. Ford (Bellevue, WA); Kevin Crews (Seattle, WA)

Assignee: Amazon Technologies, Inc.

H04L12/1822G10L15/22H04L12/1818H04L67/306H04M3/563H04W40/244G10L13/00G10L15/1815G10L2015/223H04M2242/30

Abstract

Claims (75)

1. A system for managing a conference meeting, the system comprising:

an electronic data store; and

one or more computer hardware processors in communication with the electronic data store, the one or more computer hardware processors configured to execute computer-executable instructions to at least:

receive a first audio signal from a first voice-enabled device;

identify a first user profile based on the first audio signal, wherein identifying the first user profile comprises performing speaker recognition on the first audio signal and using a first user voice profile;

receive a second audio signal from a second voice-enabled device;

identify the first user voice profile from the second audio signal;

generate a group of voice-enabled devices based on identifying the first user voice profile from both the first audio signal and the second audio signal, wherein the group comprises the first voice-enabled device and the second voice-enabled device, wherein the first voice-enabled device and the second voice-enabled device are in different rooms, wherein voice input from a conference call participant is received by the first voice-enabled device and the second voice-enabled device, and wherein the first voice-enabled device is associated with a first account different than a second account associated with the second voice-enabled device;

receive a third audio signal from the first voice-enabled device;

identify a voice command from the third audio signal;

determine, using the group, that the voice command was also received by the second voice-enabled device;

determine that the voice command corresponds to a command to leave a conference call associated with a meeting;

identify a first user profile based on the third audio signal, wherein identifying the first user profile comprises identifying the first user voice profile from the third audio signal;

identify an association between the first user profile and the first voice-enabled device;

select the first voice-enabled device, from the first voice-enabled device and the second voice-enabled device, based on the association between the first user profile and the first voice-enabled device; and

execute the voice command, wherein execution of the first voice command causes the first voice-enabled device to disconnect from the conference call.

2. The system of claim 1 , wherein the one or more computer hardware processors are further configured to:

receive, from the first voice-enabled device, an indication that a first beacon associated with the first user profile is received by the first voice-enabled device, wherein the first beacon is transmitted to the first voice-enabled device via at least one of radio-frequency identification (RFID) or Bluetooth, wherein the presence of the first beacon indicates that the user of the first user profile is near the first voice-enabled device.

3. The system of claim 1 , wherein the one or more computer hardware processors are further configured to:

retrieve event data associated with the meeting;

retrieve, from the event data, a first user profile identifier; and

retrieve the first user profile using the first user profile identifier, wherein a first user associated with the first user profile is invited to the meeting.

4. The system of claim 1 , wherein performing speaker recognition on the first audio signal comprises:

comparing the first audio signal to a baseline voice model, wherein comparing the first audio signal to the baseline model identifies a first difference between the first audio signal and the baseline model; and

selecting, from a plurality of user voice profiles, the first user voice profile, wherein selecting the first user voice profile further comprises:

identifying that the first difference is within a first threshold of a first feature of the first user voice profile.

5. A computer-implemented method comprising:

receiving a first audio signal from a first voice-enabled device;

identifying a first user profile based on the first audio signal, wherein identifying the first user profile comprises performing speaker recognition on the first audio signal and using a first user voice profile;

receiving a second audio signal from a second voice-enabled device;

identifying the first user voice profile from the second audio signal;

generating a group of voice-enabled devices based on identifying the first user voice profile from both the first audio signal and the second audio signal, wherein the group comprises the first voice-enabled device and the second voice-enabled device, wherein voice input from a conference call participant is received by the first voice-enabled device and the second voice-enabled device, and wherein the first voice-enabled device is associated with a first account different than a second account associated with the second voice-enabled device;

receiving a third audio signal from the first voice-enabled device;

identifying a voice command from a third audio signal;

determining, using the group, that the voice command was also received by the second voice-enabled device;

identifying a first user profile based on the third audio signal, wherein identifying the first user profile comprises identifying the first user voice profile from the third audio signal;

determining that the first user profile is associated with the first voice-enabled device;

selecting the first voice-enabled device, from the first voice-enabled device and the second voice-enabled device, based on the determining that the first user profile is associated with the first voice-enabled device; and

executing the voice command for the first voice-enabled device instead of the second voice-enabled device.

6. The computer-implemented method of claim 5 , further comprising:

receiving, from the first voice-enabled device, an indication that a first beacon associated with the first user profile is received by the first voice-enabled device.

7. The computer-implemented method of claim 6 , wherein the first beacon is transmitted to the first voice-enabled device via radio-frequency identification (RFID) or Bluetooth.

8. The computer-implemented method of claim 5 , further comprising:

determining that both the first audio signal and the second audio signal were received within a threshold period of time.

9. The computer-implemented method of claim 5 , wherein identifying the first user profile based on the first audio signal comprises:

identifying the first user voice profile from the first audio signal, wherein the first user voice profile comprises a feature of a first user's speech utterance; and

accessing a second entry that indicates an association between the first user voice profile and the first user profile.

10. The computer-implemented method of claim 5 , wherein the first user profile comprises a first entry, and wherein the first entry further indicates that the first voice-enabled device is registered to the first user profile.

11. A system comprising:

an electronic data store; and

one or more computer hardware processors in communication with the electronic data store, the one or more computer hardware processors configured to execute computer-executable instructions to at least:

receive a first audio signal from a first voice-enabled device;

identify a first user profile based on the first audio signal, wherein identifying the first user profile comprises performing speaker recognition on the first audio signal and using a first user voice profile;

receive a second audio signal from a second voice-enabled device;

identify the first user voice profile from the second audio signal;

generate a group of voice-enabled devices based on identifying the first user voice profile from both the first audio signal and the second audio signal, wherein the group comprises the first voice-enabled device and the second voice-enabled device, wherein voice input from a conference call participant is received by the first voice-enabled device and the second voice-enabled device, and wherein the first voice-enabled device is associated with a first account different than a second account associated with the second voice-enabled device;

receive a third audio signal from the first voice-enabled device;

identify a voice command from the third audio signal;

determine, using the group, that the voice command was also received by the second voice-enabled device;

identify a first user profile based on the third audio signal;

determine that the first user profile is associated with the first voice-enabled device;

select the first voice-enabled device, from the first voice-enabled device and the second voice-enabled device, based at least on determining that the first user profile is associated with the first voice-enabled device; and

execute the voice command for the first voice-enabled device instead of the second voice-enabled device.

12. The system of claim 11 , wherein the one or more computer hardware processors are further configured to:

receive, from the first voice-enabled device, an indication that a first beacon associated with the first user profile is received by the first voice-enabled device.

13. The system of claim 12 , wherein the first voice-enabled device is located in a room with a speaker device separate from the first voice-enabled device, and wherein audio for a conference call session is provided in the room by the speaker device.

14. The system of claim 11 , wherein the computer hardware processor is further configured to at least:

retrieve event data associated with a meeting;

retrieve the first user profile based on information in the event data.

15. The system of claim 11 , wherein identifying the first user profile based on the first audio signal comprises:

identifying the first user voice profile from the first audio signal, wherein the first user voice profile comprises a feature of a first user's speech utterance, the first user voice profile associated with the first user profile.

16. The system of claim 15 , wherein the one or more computer hardware processors are further configured to:

prompt a user to conduct a voice registration process;

receive voice input from the voice registration process; and

generate the first user voice profile based on the voice input.

Assignments (1)

Cited By (9)

US 12,211,515 US 12,284,048 US 12,334,078 US 12,475,883 US 12,587,576 US 12,602,446 US 12,652,212 US 12,687,624 US 12,707,245