Granted Patent B2

US 12,417,768 · App. 18/108,859 · Granted Sep 16, 2025

Inventors: Suneung Park (Suwon-si, KR); Sanghee Kim (Suwon-si, KR); Hyunju Seo (Suwon-si, KR)

Assignee: SAMSUNG ELECTRONICS CO., LTD.

G10L15/22G10L15/1815G10L15/30H04R3/00G10L2015/088G10L2015/223

Abstract

Claims (56)

1. An electronic device comprising:

a microphone configured to receive a first utterance of a user and a second utterance of the user, the first utterance comprising a wakeup word and additional words after the wakeup word, and the second utterance being generated at a later point in time than the first utterance;

a memory configured to store instructions; and

at least one processor configured to execute the instructions to:

extract a first domain corresponding to the first utterance and a first intent corresponding to the first utterance based on the first utterance;

execute an action corresponding to the additional words of the first utterance;

provide a response to the user regarding a result of executing the action corresponding to the additional words of the first utterance;

determine, based on the first domain and the first intent, a first correlation between the first utterance and the second utterance; and

determine, based on the first correlation, whether to process the second utterance,

wherein the second utterance does not comprise the wakeup word, and the second utterance is received from the user after the response is provided.

2. The electronic device of claim 1 , wherein the at least one processor is further configured to execute the instructions to:

based on a result of executing an action corresponding to the first domain and the first intent, generate a context corresponding to the first utterance; and

generate, based on the context, an intent set corresponding to the first correlation.

3. The electronic device of claim 2 , wherein the at least one processor is further configured to execute the instructions to:

determine a second correlation between the context and a second domain corresponding to the second utterance based on one of a degree of relationship between pre-trained domains or a usage history of the user.

4. The electronic device of claim 1 , wherein the at least one processor is further configured to execute the instructions to:

extract a second domain corresponding to the second utterance; and

determine the first correlation based on a context corresponding to the first utterance and the second domain.

5. The electronic device of claim 2 , wherein the at least one processor is further configured to execute the instructions to generate the intent set based on a predefined continuous utterance support policy.

6. The electronic device of claim 2 , wherein the at least one processor is further configured to execute the instructions to update the intent set based on a determination of whether to process the second utterance.

7. The electronic device of claim 2 , wherein the at least one processor is further configured to execute the instructions to generate the intent set using an intent comprised in the first domain and an intent comprised in a third domain different from the first domain.

8. The electronic device of claim 1 , wherein the at least one processor is further configured to execute the instructions to determine whether to activate the microphone based on a determination of whether to process the second utterance.

9. The electronic device of claim 1 , wherein the at least one processor is further configured to execute the instructions to:

based on determining that the second utterance is to be processed, activate the microphone for a predetermined time; and

based on determining that the second utterance is not to be processed, deactivate the microphone.

10. An electronic device comprising:

a microphone configured to receive a first utterance of a user and a second utterance of the user, the first utterance comprising a wakeup word and additional words after the wakeup word, and the second utterance being generated at a later point in time than the first utterance;

a memory storing instructions; and

at least one processor configured to execute the instructions to:

execute an action corresponding to the additional words of the first utterance;

provide a response to the user regarding a result of executing the action corresponding to the additional words of the first utterance;

receive a correlation list correlated with a first domain corresponding to the first utterance;

determine a first correlation between a second domain corresponding to the second utterance and the first utterance based on the correlation list; and

determine whether to process the second utterance based on a result of determining the first correlation,

wherein the second utterance does not comprise the wakeup word, and the second utterance is received from the user after the response is provided.

11. The electronic device of claim 10 , wherein the correlation list comprises a plurality of domains or plurality of intents based on the first correlation.

12. The electronic device of claim 10 , wherein the at least one processor is further configured to execute the instructions to:

based on a result of executing an action corresponding to the first domain and a first intent corresponding to the first utterance, generate a context corresponding to the first utterance; and

determine the first correlation based on the correlation list and the context.

13. The electronic device of claim 10 , wherein the at least one processor is further configured to execute the instructions to:

extract the second domain corresponding to the second utterance; and

determine the first correlation based on the correlation list and the second domain.

14. The electronic device of claim 10 , wherein the at least one processor is further configured to execute the instructions to update the correlation list based on a determination of whether to process the second utterance.

15. The electronic device of claim 10 , wherein the correlation list is generated based on one of a degree of relationship between pre-trained domains or a usage history of the user.

16. The electronic device of claim 10 , wherein the at least one processor is further configured to execute the instructions to determine whether to activate the microphone based on a result of determining whether to process the second utterance.

17. The electronic device of claim 10 , wherein the at least one processor is further configured to execute the instructions to:

based on determining that the second utterance is to be processed, activate the microphone for a predetermined time; and

based on determining that the second utterance is not to be processed, deactivate the microphone.

18. An utterance processing method of an electronic device, the utterance processing method comprising:

receiving a first utterance of a user and a second utterance of the user, the first utterance comprising a wakeup word and additional words after the wakeup word, and the second utterance being generated at a later point in time than the first utterance;

extracting, based on the first utterance, a first domain corresponding to the first utterance and a first intent corresponding to the first utterance;

executing an action corresponding to the additional words of the first utterance;

providing a response to the user regarding a result of executing the action corresponding to the additional words of the first utterance;

determining, based on the first domain and the first intent, a correlation between the first utterance and the second utterance; and

determining, based on the correlation, whether to process the second utterance,

wherein the second utterance does not comprise the wakeup word, and the second utterance is received from the user after the response is provided.

Assignments (1)

Priority Claims (2)

Continuity (2)

References Cited (62)