公开(公告)号：US10916240B2

公开(公告)日：2021-02-09

申请号：US16401726

申请日：2019-05-02

Applicant: LG ELECTRONICS INC.

Inventor： Jeehye Lee ,  Hwansik Yun ,  Eugene Kim

IPC: G10L15/14 ,  G10L15/06 ,  G06N3/08 ,  G10L15/16

Abstract: A terminal includes a memory configured to store voice data and a processor configured to measure reliability of learnable data stored in the memory, to classify the learnable data into learning data or adaptive data according to the measured reliability, to generate a learning model by performing unsupervised learning with respect to the learning data, to generate an adaptive model using the adaptive data, and to evaluate recognition performance of each of the learning model and the adaptive model.

公开(公告)号：US11551662B2

公开(公告)日：2023-01-10

申请号：US17088480

申请日：2020-11-03

Applicant: LG ELECTRONICS INC.

Inventor： Hwansik Yun ,  Wonho Shin ,  Yongchul Park ,  Sungmin Han ,  Siyoung Yang ,  Sangki Kim ,  Juyeong Jang ,  Minook Kim

IPC: G10L17/00 ,  G10L13/02 ,  G10L15/16

Abstract: A voice recognition device and a method for learning voice data using the same are disclosed. The voice recognition device combines feature information for various speakers with a text-to-speech function to generate voice data recognized by a voice recognition unit, and can improve voice recognition efficiency by allowing the voice recognition unit itself to learn various voice data. The voice recognition device can be associated with an artificial intelligence module, a robot, an augmented reality (AR) device, a virtual reality (VR) device, devices related to 5G services, and the like.

公开(公告)号：US11322134B2

公开(公告)日：2022-05-03

申请号：US16855872

申请日：2020-04-22

Applicant: LG ELECTRONICS INC.

Inventor： Hwansik Yun

IPC: G10L15/00 ,  G10L15/14 ,  G10L15/06 ,  G10L15/32

Abstract: An artificial intelligence (AI) device may acquire a probability that a received speech signal is classified as a noise signal, calculate a confidence level of a first model for determining to which phoneme the speech signal belongs, based on the speech signal, determine a weight of the first model based on the probability and the confidence level of the first model, and output a speech recognition result of the speech signal using the determined weight of the first model.