# US 12,118,983 B2 · Electronic device and operation method thereof

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,118,983 B2
- Document ID: US12118983B2
- Status: Granted
- Kind: B2
- Application number: 17/711,420
- Filed: 2022-04-01
- Published: 2022-10-06
- Granted: 2024-10-15
- Earliest claimed: 2021-04-02
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Yeona Hong; Jungwook Park; Minseok Kwon; Jungin Lee; Seokyeong Jung; Gajin Song; Jaeyung Yeo
- Web page: https://patentgrep.com/patent/US12118983B2

## CPC

- G06F 40/30: Semantic analysis
- G10L 15/063: {Training}
- G06F 3/16: Sound input; Sound output (speech processing G10L)
- G06F 40/268: Morphological analysis
- G10L 15/07: to the speaker
- G10L 15/183: using context dependencies, e.g. language models
- G10L 17/04: Training, enrolment or model building
- G06F 40/216: using statistical methods
- G06F 40/279: Recognition of textual entities
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Disclosed is an electronic device. The electronic device includes: a processor, and a memory operatively connected to the processor, the memory stores instructions that, when executed, cause the processor to: select at least one data received through a user input, analyze the selected data, extract additional data based on the analyzed data, learn a personal voice model using the data and the additional data, and provide response data using the personal voice model.

## Claims

### Claim 1

An electronic device comprising:
a processor; and
a memory operatively connected to the processor,
wherein the memory stores instructions that, when executed, cause the processor to:
select at least one data received through a user input;
analyze text data included in the selected at least one data using morphological analysis;
identify a substantive portion contained in the text data using the morphological analysis;
embed the identified substantive portion contained in the text data;
extract additional data including at least one of a word and a phrase having a greatest similarity with the identified substantive portion;
learn an external language model using the selected at least one data;
learn a personal language model using the selected at least one data and the additional data;
adjust a first weight value for a learning result of the personal language model to be higher than a second weight value for a learning result of the external language model, based on the additional data being extracted; and
provide response data using the learning result of the personal language model based on the adjusted first weight value.

### Claim 2 (depends on 1)

The electronic device of claim 1, further comprising:
wherein the instructions, when executed, cause the processor to process a voice signal contained in the user input using a sound model to extract the text data.

### Claim 3 (depends on 1)

The electronic device of claim 1, wherein the instructions, when executed, cause the processor to analyze image data included in the selected at least one data using an optical character reader.

### Claim 4 (depends on 3)

The electronic device of claim 3, wherein the instructions, when executed, cause the processor to distinguish texts and images contained in the image data from each other using the optical character reader.

### Claim 5 (depends on 4)

The electronic device of claim 4, wherein the instructions, when executed, cause the processor to embed the distinguished images to extract at least one of images with a greatest similarity as the additional data.

### Claim 6 (depends on 1)

The electronic device of claim 1, wherein the instructions, when executed, cause the processor to learn the personal language model independently from the external language model.

### Claim 7

A method performed by an electronic device, the method comprising:
selecting at least one data received through a user input;
analyzing text data included in the selected at least one data using morphological analysis;
identifying a substantive portion contained in the text data using the morphological analysis;
embedding the identified substantive portion contained in the text data;
extracting additional data including at least one of a word and a phrase having a greatest similarity with the identified substantive portion;
learning an external language model using the selected at least one data;
learning a personal language model using the selected at least one data and the additional data;
adjusting a first weight value for a learning result of the personal language model to be higher than a second weight value for a learning result of the external language model, based on the additional data being extracted; and
providing response data using the learning result of the personal language model based on the adjusted first weight value.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
receiving a voice signal contained in the user input using a sound model included in the electronic device; and
training the sound model using a learning algorithm.

### Claim 9 (depends on 7)

The method of claim 7, further comprising:
analyzing image data included in the selected at least one data using an optical character reader.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
distinguishing texts and images contained in the image data from each other using the optical character reader.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
embedding the distinguished images to extract at least one of images with a greatest similarity as the additional data.

### Claim 12 (depends on 7)

The method of claim 7, further comprising: learning the personal language model independently from the external language model.