Method and apparatus for artificial intelligence model personalization

1. An electronic apparatus comprising:

a touch display;

a microphone;

a speaker;

a memory configured to store a plurality of training data generation models and an artificial intelligence model, wherein the plurality of training data generation models comprises a personal training data generation model trained based on personal use data to generate personal training data that reflects characteristics of a user, and a general training data generation model trained to generate general training data corresponding to general use data of a plurality of users; and

a processor configured to:

control the touch display or the speaker to provide guide information requesting the personal use data;

obtain the personal use data through the touch display or the microphone;

generate the personal training data that reflects a characteristic of the user using the personal use data and the personal training data generation model;

generate the general training data that reflects a characteristic of the plurality of users;

identify whether a learning start condition is satisfied based on the artificial intelligence model incorrectly recognizing input data received through the touch display or the microphone a predetermined number of times;

train the artificial intelligence model using the personal training data and the general training data as training data, based on the learning start condition being satisfied;

store the artificial intelligence model trained using the personal training data and the general training data in the memory; and

perform recognition of input data received through the touch display or the microphone using the artificial intelligence model trained using the personal training data and the general training data in the memory.

2. The electronic apparatus as claimed in claim 1 , wherein the processor is further configured to update the artificial intelligence model based on at least one of the personal training data, the general training data, or the personal use data.

3. The electronic apparatus as claimed in claim 2 , wherein the processor is further configured to update the personal training data generation model based on at least one of the personal use data or the personal training data.

4. The electronic apparatus as claimed in claim 1 , wherein the artificial intelligence model is a voice recognition model, a handwriting recognition model, an object recognition model, a speaker recognition model, a word recommendation model, or a translation model.

5. The electronic apparatus as claimed in claim 2 , wherein the general training data includes first input data,

the personal training data includes second input data, and

the artificial intelligence model performs unsupervised learning based on the personal use data, the first input data, and the second input data.

6. The electronic apparatus as claimed in claim 2 , wherein the general training data includes first input data,

the personal training data includes second input data, and

the artificial intelligence model generates first output data corresponding to the first input data based on the first input data being input, generates second output data corresponding to the second input data based on the second input data being input, and is trained based on the personal use data, the first input data, the first output data, the second input data, and the second output data.

7. The electronic apparatus as claimed in claim 1 , wherein the general training data generation model is downloaded from a server and stored in the memory.

8. The electronic apparatus as claimed in claim 7 , wherein the processor is configured to upload the artificial intelligence model to the server.

9. The electronic apparatus as claimed in claim 1 , wherein the processor is configured to train the artificial intelligence model based on the electronic apparatus being in a charge state, an occurrence of a predetermined time, or detecting no manipulation of the electronic apparatus by the user for a predetermined time.

10. A control method of an electronic apparatus including a touch display, a microphone, and speaker, a plurality of training data generation models and an artificial intelligence model, wherein the plurality of training data generation models includes a personal training data generation model trained based on personal use data to generate personal training data that reflects characteristics of a user and a general training data generation model trained to generate general training data corresponding to general use data of a plurality of users, the control method comprising:

controlling the touch display or the speaker to provide guide information requesting the personal use data;

obtaining the personal use data through the touch display or the microphone;

generating the personal training data that reflects a characteristic of the user using the personal use data and the personal training data generation model;

generating the general training data that reflects a characteristic of the plurality of users;

identifying whether a learning start condition is satisfied based on dthe artificial intelligence model incorrectly recognizing input data received through the touch display or the microphone a predetermined number of times;

training the artificial intelligence model using the personal training data and the general training data as training data, based on the learning start condition being satisfied;

storing the artificial intelligence model trained using the personal training data and the general training data; and

performing recognition of input data received through the touch display or the microphone using the artificial intelligence model trained using the personal training data and the general training data.

11. The control method as claimed in claim 10 , wherein the artificial intelligence model is a model that is updated based on at least one of the personal training data, the general training data, or the personal use data.

12. The control method as claimed in claim 11 , wherein the personal training data generation model is updated based on at least one of the personal use data or the personal training data.

13. The control method as claimed in claim 10 , wherein the artificial intelligence model is a voice recognition model, a handwriting recognition model, an object recognition model, a speaker recognition model, a word recommendation model, or a translation model.

14. The control method as claimed in claim 11 , wherein

the general training data includes first input data,

the personal training data includes second input data, and the artificial intelligence model performs unsupervised learning based on the personal use data, the first input data, and the second input data.

15. The control method as claimed in claim 11 , wherein the general training data includes first input data,

the personal training data include second input data, and

the artificial intelligence model generates first output data corresponding to the first input data based on the first input data being input, generates second output data corresponding to the second input data based on the second input data being input, and is trained based on the personal use data, the first input data, the first output data, the second input data, and the second output data.

16. The control method as claimed in claim 10 , wherein the general training data generation model is downloaded from a server and stored in a memory of the electronic apparatus.

17. The control method as claimed in claim 16 , wherein the artificial intelligence model is uploaded to the server.

18. The control method as claimed in claim 10 , further comprising training the artificial intelligence model based on the electronic apparatus being in a charge state, an occurrence of a predetermined time, or detecting no manipulation of the electronic apparatus by the user for a predetermined time.

19. A non-transitory computer-readable medium configured to store a program for performing a method for personalization of an artificial intelligence model in an electronic apparatus using a plurality of training data generation models, wherein the plurality of training data generation models includes a personal training data generation model trained based on personal use data to generate personal training data that reflects characteristics of a user and a general training data generation model trained to generate general training data corresponding to general use data of a plurality of users, the method including:

controlling a touch display or a speaker to provide guide information requesting the personal use data;

obtaining the personal use data through the touch display or a microphone;

generating the personal training data that reflects a characteristic of the user using the personal use data and the personal training data generation model;

generating the general training data that reflects a characteristic of the plurality of users;

identifying whether a learning start condition is satisfied based on the artificial intelligence model incorrectly recognizing input data received through the touch display or the microphone a predetermined number of times;

training the artificial intelligence model using the personal training data and the general training data as training data, based on the learning start condition being satisfied; and

performing recognition of input data received through the touch display or the microphone using the artificial intelligence model trained using the personal training data and the general training data.

20. The non-transitory computer-readable medium as claimed in claim 19 , wherein the method further includes training the artificial intelligence model using at least one of the personal training data, the general training data, or actual use data of the user.