Electronic device and method for controlling the electronic device

1. An electronic device comprising:

a microphone;

a non-volatile memory configured to store virtual assistant model data comprising data that is classified according to a plurality of domains and data that is commonly used for the plurality of domains;

a volatile memory; and

a processor configured to:

based on receiving, through the microphone, a trigger input to perform speech recognition for a user speech, initiate loading the virtual assistant model data from the non-volatile memory into the volatile memory,

load, into the volatile memory, first data corresponding to at least one first domain regarding a first user speech received through the microphone from among the data classified according to the plurality of domains, and

while loading the first data into the volatile memory, load at least a part of the data commonly used for the plurality of domains into the volatile memory.

2. The electronic device of claim 1 , wherein the processor is further configured to:

based on receiving the first user speech through the microphone, obtain a first text corresponding to the first user speech by performing speech recognition for the first user speech; and

based on the obtained first text, identify the at least one first domain corresponding to the first user speech among the plurality of domains.

3. The electronic device of claim 2 , wherein the processor is further configured to, based on the identified at least one first domain, load, as the first data, data corresponding to the identified at least one first domain, from among the data classified according to the plurality of domains, into the volatile memory.

4. The electronic device of claim 3 , wherein the processor is further configured to:

process the loaded first data corresponding to the identified at least first one domain; and

perform natural language understanding for the first text based on the processed first data.

5. The electronic device of claim 3 , wherein the processor is further configured to, maintain the loaded first data corresponding to the identified at least one first domain in the volatile memory for a predetermined period.

6. The electronic device of claim 5 , wherein the processor is further configured to:

based on receiving a second user speech through the microphone while the first data corresponding to the identified at least one first domain is maintained in the volatile memory, identify at least one second domain corresponding to the second user speech; and

load second data, among the data classified according to the plurality of domains, corresponding to the identified at least one second domain and different from the first data, into the volatile memory.

7. The electronic device of claim 1 , wherein the processor is further configured to:

based on the electronic device being turned on, load data, among the data commonly used for the plurality of domains, corresponding to a module for recognizing the trigger input, into the volatile memory; and

based on recognizing the trigger input using the loaded data corresponding to the module for recognizing the trigger input, determine to perform the speech recognition for the user speech.

8. A method for controlling an electronic device comprising a volatile memory and a non-volatile memory storing virtual assistant model data including data classified in accordance with a plurality of domains and data commonly used for the plurality of domains, the method comprising:

based on receiving a trigger input to perform speech recognition for a user speech, initiating loading the virtual assistant model data from the non-volatile memory into the volatile memory;

loading, into the volatile memory, first data corresponding to at least one first domain regarding a first user speech received through the microphone, from among the data classified according to the plurality of domains; and

while loading the first data into the volatile memory, loading at least a part of the data commonly used for the plurality of domains into the volatile memory.

9. The method of claim 8 , further comprising:

based on receiving the first user speech, obtaining a first text corresponding to the first user speech by performing speech recognition for the first user speech; and

based on the obtained first text, identifying the at least one first domain corresponding to the first user speech among the plurality of domains.

10. The method of claim 9 , wherein the loading the first data comprises:

based on the identified at least one first domain, loading, as the first data, data corresponding to the identified at least one first domain, from among the data classified according to the plurality of domains, into the volatile memory.

11. The method of claim 10 , further comprising:

processing the loaded first data corresponding to the identified at least one first domain; and

performing natural language understanding for the first text based on the processed first data.

12. The method of claim 10 , further comprising:

maintaining the loaded first data corresponding to the identified at least one first domain in the volatile memory for a predetermined period.

13. The method of claim 12 , further comprising:

based on receiving a second user speech while the first data corresponding to the identified at least one first domain is maintained in the volatile memory, identifying at least one second domain corresponding to the second user speech; and

loading second data, among the data classified according to the plurality of domains, corresponding to the identified at least one second domain and different from the first data, into the volatile memory.

14. The method of claim 8 , further comprising:

based on the electronic device being turned on, loading data, among the data commonly used for the plurality of domains, corresponding to a module for recognizing the trigger input, into the volatile memory; and

based on recognizing the trigger input using the loaded data corresponding to the module for recognizing the trigger input, determining to perform the speech recognition for the user speech.

15. A non-transitory computer readable recording medium having recorded thereon instructions executable by at least one processor for performing the method of claim 8 .

16. An electronic device comprising:

a non-volatile memory configured to store virtual assistant model data comprising data that is classified according to a plurality of domains and data that is commonly used for the plurality of domains;

a volatile memory; and

a processor configured to:

based on determining that a trigger input to perform speech recognition for a user speech is received, loading, into the volatile memory, first data corresponding to at least one first domain regarding a first user speech received through the microphone from among the data classified according to the plurality of domains, and

while loading the first data into the volatile memory, load at least a part of the data commonly used for the plurality of domains into the volatile memory.

17. The electronic device of claim 16 , wherein the processor is further configured to:

based on receiving the first user speech, obtain a first text corresponding to the first user speech by performing speech recognition for the first user speech; and

based on the obtained first text, identify the at least one first domain corresponding to the first user speech among the plurality of domains.

18. The electronic device of claim 17 , wherein the processor is further configured to, based on the identified at least one first domain, load, as the first data, data corresponding to the identified at least one first domain, from among the data classified according to the plurality of domains, into the volatile memory.

19. The electronic device of claim 18 , wherein the processor is further configured to:

process the loaded first data corresponding to the identified at least first one domain; and

perform natural language understanding for the first text based on the processed first data.

20. The electronic device of claim 18 , wherein the processor is further configured to:

based on receiving a second user speech while the first data corresponding to the identified at least one first domain is maintained in the volatile memory, identify at least one second domain corresponding to the second user speech; and

load second data, among the data classified according to the plurality of domains, corresponding to the identified at least one second domain and different from the first data, into the volatile memory.