Patent · US Active

Speech recognition method and apparatus, device, and storage medium

US12230250B2 · kind B2 · utility

0Cited by

2References

20Claims

0Family size

Assignee

TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED · CN

Inventors

Jun Wang · Lo Wu, CN
Wing Yip Lam · Hong Kong, CN

Key dates

Filing date	Feb 14, 2022
Grant date	Feb 18, 2025
Priority date	—
Expiry date	Feb 10, 2043

Classification

Technology area (CPC G)Physics
CPC primaryG10L2015/0635
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A speech recognition method includes: obtaining first sample speech data corresponding to a target user and a first reference speech recognition result corresponding to the first sample speech data; obtaining a pre-update target model; inputting the first sample speech data into the pre-update target model, and performing speech recognition by using a target speech extraction model, a target feature extraction model, and a target speech recognition model, to obtain a first model output result; obtaining a target model loss value corresponding to the target feature extraction model according to the first model output result and the first reference speech recognition result; and updating a model parameter of the target feature extraction model in the pre-update target model according to the target model loss value, to obtain a post-update target model.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.