Method of providing intelligent voice recognition model for voice recognition device

Patent metadata

Field	Value
Publication number	US-11849908-B2
Application number	US-201916489952-A
Country	US
Kind code	B2
Filing date	Jun 5, 2019
Priority date	Jun 5, 2019
Publication date	Dec 26, 2023
Grant date	Dec 26, 2023

Abstract

A method of providing an intelligent voice recognition model includes obtaining space type information about a placement area of the voice recognition device, extracting space feature information from the space type information; and generating a predetermined voice recognition model matched to the extracted space feature information. At least one device implementing the method of providing the intelligent voice recognition model may be associated with an artificial intelligence module, a unmanned aerial vehicle (UAV), a robot, an augmented reality (AR) device, a virtual reality (VR) device, devices related to 5G services, and the like.

First claim

The invention claimed is:

1. A method of providing an intelligent voice recognition model for a voice recognition device, the method comprising:
obtaining space type information about a placement area of the voice recognition device;
extracting space feature information from the space type information; and
generating a predetermined voice recognition model matched to the extracted space feature information.

2. The method of claim 1 , wherein the obtaining of the space type information comprises:
obtaining sensing information corresponding to a specific physical quantity or analog information of the placement area using a sensing device; and
obtaining the space type information based on the sensing information.

3. The method of claim 2 , wherein the obtaining of the sensing information is performed using the sensing device mounted on the voice recognition device.

4. The method of claim 2 , wherein the obtaining of the sensing information is performed using the sensing device mounted on a mobile agent that is separate from the voice recognition device.

5. The method of claim 2 , wherein the obtaining of the sensing information comprises obtaining 2D or 3D image information of the placement area.

6. The method of claim 5 , wherein the obtaining of the space type information comprises obtaining area information of the placement area based on the image information.

7. The method of claim 5 , wherein the obtaining of the space type information comprises obtaining distance information between the voice recognition device and boundary surfaces of the placement area based on the image information.

8. The method of claim 2 , wherein the obtaining of the sensing information comprises obtaining acoustic information of the placement area.

9. The method of claim 8 , wherein the obtaining of the space type information comprises obtaining noise information of the placement area based on the acoustic information.

10. The method of claim 2 , wherein the obtaining of the space type information comprises:
transmitting the sensing information to a server; and
obtaining, by the server, the space type information based on the sensing information.

11. The method of claim 10 , further comprising:
obtaining, by the voice recognition device, voice information;
receiving, by the server, the voice information from the voice recognition device;
generating, by the server, text information corresponding to the voice information; and
transmitting, by the server, the text information to the voice recognition device.

12. The method of claim 11 , wherein the receiving of the voice information comprises receiving an identification code distinguishing the voice recognition device.

13. The method of claim 2 , wherein the obtaining of the space type information comprises obtaining first space type information and second space type information.

14. The method of claim 13 , wherein the obtaining of the space type information comprises combining the first space type information and the second space type information to generate combined space type information,
wherein the extracting of the space feature information comprises searching for the space feature information matched to the combined space type information that is previously set.

15. The method of claim 13 , wherein the generating of the voice recognition model comprises:
searching for a first voice recognition model matched to the first space type information at a predetermined first voice recognition model table;
searching for a second voice recognition model matched to the second space type information at a predetermined second voice recognition model table; and
combining the first and second voice recognition models to generate the voice recognition model.

Assignees

• Lg Electronics Inc

Inventors

• Chae Jonghoon

Classifications

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G10L15/30

  Distributed recognition, e.g. in client-server systems, for mobile phones or network applications · CPC title

• A47L9/2805 Primary

  Parameters or conditions being sensed · CPC title

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title