Granted Patent B2

US 12,417,764 · App. 17/248,859 · Granted Sep 16, 2025

Inventors: Jaesick Shin (Suwon-si, KR); Sungho Lee (Suwon-si, KR)

Assignee: Samsung Electronics Co., Ltd.

G10L15/22G10L15/16G10L15/18G10L15/26

Abstract

Claims (88)

1. A method for providing a voice assistant service, the method comprising:

receiving a first voice command from a user;

determining, from among a plurality of candidate devices, a plurality of devices for the first voice command;

transmitting, to the plurality of devices, information related to the first voice command;

receiving, from the plurality of devices, a plurality of service provision messages generated in response to receiving the first voice command and a plurality of pieces of service provision history information of the plurality of devices, wherein the service provision history information includes time information indicating a current time when a service is selected in response to receiving the first voice command,

identifying a type of the service requested by the first voice command, comprising a first type and a second type, wherein the plurality of devices is prioritized based on the time information indicating the current time and location information of the plurality of devices when the type of the service is identified as the first type and the plurality of devices is prioritized based on device information when the type of the service is identified as the second type;

selecting, based on the plurality of the pieces of service provision history information, at least one of the plurality of service provision messages; and

outputting, based on a selection of the at least one of the plurality of service provision messages, a response message in response to receiving the first voice command.

2. The method of claim 1 , wherein determining the plurality of devices comprises:

converting the first voice command into a first text;

interpreting the first text by using a natural language understanding (NLU) model;

determining an intent of the user based on a result of the interpreting of the first text; and

determining, based on a relevance between the intent of the user and each of the plurality of candidate devices, the plurality of devices for the first voice command, from among the plurality of candidate devices.

3. The method of claim 2 , wherein determining the plurality of devices based on the relevance comprises:

obtaining a plurality of pieces of the device information, respectively, regarding the plurality of candidate devices;

obtaining, based on the plurality of pieces of the device information, a plurality of probability values indicating a degree of the relevance between the intent of the user and each of the plurality of candidate devices; and

determining the plurality of devices including probability values that are greater than or equal to a threshold value.

4. The method of claim 1 , wherein a first service provision message received from a first device among the plurality of devices includes at least one of information whether the first device is able to provide the service in response to receiving the first voice command, information whether the first device is included in a first group, identification information of the first device, identification information of a first service provided by the first device in response to receiving the first voice command, a type of the first service, or identification information of an application used to provide the first service.

5. The method of claim 4 , wherein selecting the at least one of the plurality of service provision messages comprises:

identifying, from among the plurality of devices, one or more devices included in the first group; and

selecting, based on the plurality of pieces of service provision history information of the identified one or more devices, at least one service provision message from among the plurality of service provision messages received from the identified one or more devices.

6. The method of claim 1 , wherein each of the plurality of devices is capable of providing the service based on information for sub-devices including at least one of a screen or an input interface in response to receiving the first voice command.

7. The method of claim 1 , wherein:

receiving the plurality of service provision messages and the plurality of pieces of service provision history information comprises, respectively, receiving, from the plurality of devices, a plurality of pieces of the device information, each piece of the device information regarding sub-devices constituting a corresponding device, together with the plurality of service provision messages and the plurality of pieces of service provision history information, and

selecting the at least one of the plurality of service provision messages comprises selecting at least one of the plurality of service provision messages, based on the plurality of pieces of the device information regarding the plurality of devices and the plurality of pieces of service provision history information thereof.

8. The method of claim 1 , wherein selecting the at least one of the plurality of service provision messages comprises selecting at least one of the plurality of service provision messages by using a service recommendation model, and

wherein the service recommendation model comprises:

an artificial intelligence (AI) algorithm trained based on a voice command,

service provision histories for the plurality of devices with respect to the voice command, and

a plurality of pieces of the device information regarding the plurality of devices.

9. The method of claim 1 , wherein selecting the at least one of the plurality of service provision messages comprises:

assigning priorities to the plurality of service provision messages based on at least one of information whether each device provides the service in response to receiving the first voice command, the type of the service provided by each device in response to receiving the first voice command, a number of times that the service provided by each device in response to receiving the first voice command was selected, context information when the service was selected in response to receiving the first voice command, or the device information regarding sub-devices constituting each device; and

selecting at least one of the plurality of service provision messages, based on the assigned priorities.

10. The method of claim 1 , wherein selecting the at least one of the plurality of service provision messages comprises:

assigning priorities to the plurality of service provision messages based on at least one of information whether each device provides the service in response to receiving the first voice command, the type of the service provided by each device in response to receiving the first voice command, a number of times that the service provided by each device in response to receiving the first voice command was selected, context information obtained when the service provided by each service in response to receiving the first voice command was selected, or the device information regarding sub-devices constituting each device; and

selecting, based on the assigned priorities, two or more service provision messages from among the plurality of service provision messages.

11. The method of claim 1 , wherein:

each of the plurality of service provision messages, respectively, received from the plurality of devices includes information related to the service provided by each of the plurality of devices in response to receiving the first voice command; and

the outputting of the response message in response to receiving the first voice command comprises:

generating the response message to include information related to at least one service in the selected at least one of the plurality of service provision messages; and

outputting the response message.

12. The method of claim 1 , wherein:

selecting the at least one of the plurality of service provision messages comprises:

identifying, based on the plurality of service provision messages, one or more devices included in a first group from among the plurality of devices; and

by using a service recommendation model, selecting, based on the plurality of pieces of service provision history information of the identified one or more devices, at least one service provision message from among the plurality of service provision messages received from the identified one or more devices, and

the method further comprising:

receiving a second voice command from the user;

determining, based on the second voice command, the service selected by the user from among services provided by the plurality of devices in response to receiving the first voice command;

transmitting information related to the service selected by the user to the plurality of devices; and

training the service recommendation model by using the information related to the service selected by the user.

13. An apparatus for providing a voice assistant service, the apparatus comprising:

a receiver configured to receive a command from a user;

a communicator;

memory configured to store voice assistant program including one or more instructions; and

at least one processor operably connected to the receiver, the communicator, the memory, and, where the processor is configured to:

control the receiver to receive a first voice command from the user;

determine, from among a plurality of candidate devices, a plurality of devices to that the first voice command is to be transmitted;

control the communicator to transmit information related to the first voice command to the plurality of devices and respectively receive, from the plurality of devices, a plurality of service provision messages generated in response to receiving the first voice command and a plurality of pieces of service provision history information of the plurality of devices, wherein the service provision history information includes time information indicating a current time when a service is selected in response to receiving the first voice command;

identify a type of the service requested by the first voice command, comprising a first type and a second type, wherein the plurality of devices is prioritized based on the time information indicating the current time and location information of the plurality of devices when the type of the service is identified as the first type and the plurality of devices is prioritized based on device information when the type of the service is identified as the second type;

select, based on the plurality of pieces of service provision history information, at least one of the plurality of service provision messages; and

output, based on a selection of the at least one of the plurality of service provision message, a response message in response to receiving the first voice command.

14. The apparatus of claim 13 , wherein a first service provision message received from a first device among the plurality of devices includes at least one of information whether the first device is included in a first group, identification information of the first device, identification information of a first service provided by the first device in response to receiving the first voice command, a type of the first service, or identification information of an application used to provide the first service.

15. The apparatus of claim 14 , wherein the at least one processor is further configured to:

identify, from among the plurality of devices, one or more devices included in the first group; and

select, based on the plurality of pieces of service provision history information of the identified one or more devices, at least one service provision message from among the plurality of service provision messages received from the identified one or more devices.

16. The apparatus of claim 13 , wherein service provision history information of a first device includes at least one of a number of times that a first service suggested by the first device in response to receiving the first voice command was selected by the user or context information obtained when the first service was selected in response to receiving the first voice command.

17. The apparatus of claim 13 , wherein:

the communicator is configured to respectively receive, from the plurality of devices, a plurality of pieces of the device information, each piece of the device information regarding sub-devices constituting a corresponding device and

the at least one processor is further configured to select at least one of the plurality of service provision messages based on the plurality of pieces of the device information regarding the plurality of devices and the plurality of pieces of service provision history information thereof.

18. The apparatus of claim 13 , wherein the at least one processor is further configured to select at least one of the plurality of service provision messages by using a service recommendation model, and

wherein the service recommendation model comprises an artificial intelligence (AI) algorithm trained based on a voice command, service provision histories for the plurality of devices with respect to the voice command, and a plurality of pieces of the device information regarding the plurality of devices.

19. The apparatus of claim 13 , wherein:

the at least one processor is further configured to:

identify, based on the plurality of service provision messages, one or more devices included in a first group from among the plurality of devices; and

by using a service recommendation model, select, based on the plurality of pieces of service provision history information of the identified one or more devices, at least one service provision message from among the plurality of service provision messages received from the identified one or more devices,

the receiver is further configured to further receive a second voice command; and

the at least one processor is further configured to:

determine, based on the second voice command, the service selected by the user from among services provided by the plurality of devices in response to receiving the first voice command,

control the communicator to transmit information related to the service selected by the user to the plurality of devices, and

train the service recommendation model by using the information related to the service selected by the user.

20. A non-transitory computer-readable medium embodying a computer program, the computer program comprising computer readable program code that when executed by a processor of an electronic device causes processor to:

receive a first voice command from a user;

determine, from among a plurality of candidate devices, a plurality of devices that the first voice command is to be transmitted;

transmit information related to the first voice command to the plurality of devices;

respectively receive, from the plurality of devices, a plurality of service provision messages generated in response to receiving the first voice command and a plurality of pieces of service provision history information of the plurality of devices, wherein the service provision history information includes time information indicating a current time when a service is selected in response to receiving the first voice command, and

identify a type of the service requested by the first voice command, comprising a first type and a second type, wherein the plurality of devices is prioritized based on the time information indicating the current time and location information of the plurality of devices when the type of the service is identified as the first type and the plurality of devices is prioritized based on device information when the type of the service is identified as the second type;

select, based on the plurality of pieces of service provision history information, at least one of the plurality of service provision messages; and

output, based on a selection of the at least one of the plurality of service provision messages, a response message in response to receiving the first voice command.

Assignments (1)

Priority Claims (2)

Continuity (1)

References Cited (31)