# US 2026/0064255 A1 · Method for creating a genogram using touch and voice input

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

- Number: US 2026/0064255 A1
- Document ID: US20260064255A1
- Status: Published application
- Kind: A1
- Application number: 19/074,565
- Filed: 2025-03-10
- Published: 2026-03-05
- Earliest claimed: 2024-08-28
- Assignees: National Applied Research Laboratories
- Applicants: National Applied Research Laboratories
- Inventors: Ming-Jen WANG; Chia-Chen KUO; Chia-Yung JUI; Yi-Siang LIAO; Han-Ye JI
- Web page: https://patentgrep.com/patent/US20260064255A1

## CPC

- G06F 3/04845: for image manipulation, e.g. dragging, rotation, expansion or change of colour
- G06F 3/04817: using icons (graphical or visual programming using iconic symbols G06F8/34)
- G06F 3/0488: using a touch-screen or digitiser, e.g. input of commands through traced gestures
- G06T 11/26: Drawing of charts or graphs
- G10L 15/063: {Training}
- G10L 15/16: using artificial neural networks
- G10L 15/183: using context dependencies, e.g. language models
- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L 15/30: Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
- G06T 2200/24: involving graphical user interfaces [GUIs]
- G10L 2015/0633: {using lexical or orthographic knowledge sources}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A method for creating a genogram of a family using touch and voice input includes: in response to a user touching a part of a touchscreen, generating a touch signal indicating a set of coordinates on the touchscreen at which a touch action occurred; recording a voice input from the user, the voice input including speech describing the family; obtaining an input text that is converted from the voice input; obtaining, using a generative language model based on a content of the input text, a genogram dataset in a format that is for generating the genogram; transforming the genogram dataset into a graphical genogram dataset; and creating the genogram based on the graphical genogram dataset, using the set of coordinates as a reference point, the genogram including at least one icon representing a member of the family.

## Claims

### Claim 1

A method for creating a genogram of a family using touch and voice input, the method being implemented using an electronic device that includes a touchscreen, and a cloud server that is in communication with the electronic device, the method comprising:
a) in response to a user touching a part of the touchscreen, detecting a touch action and generating a touch signal indicating a first set of coordinates on the touchscreen at which the touch action occurred;
b) activating a recording module of the electronic device to record a first voice input from the user, the first voice input including speech describing the family;
c) obtaining a first input text that is converted from the first voice input;
d) obtaining, using a generative language model based on a content of the first input text, a first genogram dataset in a format that is for generating the genogram;
e) transforming the first genogram dataset into a first graphical genogram dataset; and
f) creating a part of the genogram based on the first graphical genogram dataset, using the first set of coordinates as a reference point, the part of the genogram including at least one icon representing a member of the family.

### Claim 2 (depends on 1)

The method as claimed in claim 1, further comprising, after step f), steps of:
in response to the user touching another part of the touchscreen, detecting another touch action and generating another touch signal indicating a second set of coordinates of the touchscreen at which the touch action occurred;
activating the recording module of the electronic device to record a second voice input from the user, the second voice input including speech describing a person related to the member of the family;
obtaining a second input text that is converted from the second voice input;
obtaining, using the generative language model based on a content of the second input text, a second genogram dataset in a format that is for generating the genogram;
transforming the second genogram dataset into a second graphical genogram dataset; and
creating another part of the genogram based on the second graphical genogram dataset, using the second set of coordinates as a reference point.

### Claim 3 (depends on 2)

The method as claimed in claim 2, wherein the touch action related to the user touching the another part of the touchscreen occurred on the at least one icon, and the another part of the genogram extends from the at least one icon.

### Claim 4 (depends on 2)

The method as claimed in claim 2, wherein the person is another member of the family.

### Claim 5 (depends on 2)

The method as claimed in claim 2, wherein the person is a non-family member, and the another part of the genogram includes another icon, which is in a shape different from that of the at least one icon.

### Claim 6 (depends on 2)

The method as claimed in claim 2, wherein:
the method further comprises, prior to step a), implementing an installation process to store a number of predetermined prompts in a genogram extractor;
step d) includes the genogram extractor sending, an input prompt that is related to one of the number of predetermined prompts and that includes the first input text, to the generative language model, and the generative language model generating the first genogram dataset as a reply;
the obtaining a second genogram dataset includes the genogram extractor sending, another input prompt that is related to one of the number of predetermined prompts and that includes the second input text, to the generative language model, and the generative language model generating the second genogram dataset as a reply; and
each of the first genogram dataset and the second genogram dataset includes identification of at least one member of the family, a description of the at least one member, and a relationship of the at least one member with another member of the family.

### Claim 7 (depends on 2)

The method as claimed in claim 2, wherein each of the first genogram dataset and the second genogram dataset is in the format of JavaScript Object Notation (JSON).

### Claim 8 (depends on 1)

The method as claimed in claim 1, wherein:
step c) is implemented using a speech-to-text module; and
the method further comprises, prior to step a), a step of training a neural network model using a genogram training dataset to serve as the speech-to-text module.

### Claim 9 (depends on 8)

The method as claimed in claim 8, wherein the neural network model is a Whisper speech recognition system.

### Claim 10 (depends on 8)

The method as claimed in claim 8, wherein the training a neural network model includes a fine-tuning operation using a Low-Rank Adaptation (LoRA) technique.

### Claim 11 (depends on 1)

The method as claimed in claim 1, wherein the generative language model is embodied using Large Language Model Meta AI (LLaMA).

### Claim 12 (depends on 1)

The method as claimed in claim 1, the cloud server including a speech-to-text module, wherein:
step c) includes the electronic device transmitting the first voice input to the cloud server, to enable the speech-to-text module of the cloud server to convert the first voice input into the first input text.

### Claim 13 (depends on 1)

The method as claimed in claim 1, the cloud server including a generative language model that operates based on information of a genogram extractor, wherein:
step d) includes the genogram extractor sending an input prompt that includes the content of the first input text to the generative language model, the generative language model generating the first genogram dataset as a reply, and the cloud server transmitting the first genogram dataset to the electronic device.