Patent · US Active

Systems and methods for automatic speech recognition based on graphics processing units

US11741967B2 · kind B2 · utility

0Cited by

1References

22Claims

0Family size

Assignee

KWAI INC. · US

Inventors

Yongxiong Ren · San Jose, US
Heng Liu · Beijing, CN
Yang Liu · Nanhu, CN
Lingzhi Liu · San Jose, US
Jie Li · Lo Wu, CN
Yuanyuan Zhao · Lake Forest, US
Xiaorui Wang · Beijing, CN

Key dates

Filing date	Jan 4, 2021
Grant date	Aug 29, 2023
Priority date	—
Expiry date	Aug 29, 2041

Classification

Technology area (CPC G)Physics
CPC primaryG10L15/16
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

An automatic speech recognition system and a method thereof are provided. The system includes an encoder and a decoder. The encoder comprises a plurality of encoder layers. At least one encoder layer includes a plurality of encoder sublayers fused into one or more encoder kernels. The system further comprises a first pair of ping-pong buffers communicating with the one or more encoder kernels. The decoder comprises a plurality of decoder layers. At least one decoder layer includes a plurality of decoder sublayers fused into one or more decoder kernels. The decoder receives a decoder output related to the encoder output and generates a decoder output. The encoder sends the decoder output to a beam search kernel.

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