METHOD AND DEVICE FOR SPEECH RECOGNITION, COMPUTER DEVICE, AND ELECTRONIC DEVICE

Title:

METHOD AND DEVICE FOR SPEECH RECOGNITION, COMPUTER DEVICE, AND ELECTRONIC DEVICE

Document Type and Number:

WIPO Patent Application WO/2020/001163

Kind Code:

A1

Abstract:

The present application relates to a method and device for speech recognition, a computer device, and an electronic device. The method comprises: receiving an audio signal captured by a microphone array; performing beamforming processing of the audio signal respectively in multiple different target directions to produce multiple corresponding beamforming signals; performing speech recognition respectively with respect to each beamforming signal to produce speech recognition results of the beamforming signals; and determining a speech recognition result of the audio signal on the basis of the speech recognition results of the beamforming signals. The method obviates the need to take into consideration the direction of a sound source, with beamforming processing performed in the different target directions, at least one of the target directions comes close to the actual direction in which sound is generated; therefore, the beamforming signal in at least one of the target directions is clear when enhanced, thus allowing speech recognition to be performed on the basis of the beamforming signals, and increasing the accuracy of speech recognition.

Inventors:

GAO YI (CN)
ZHENG JIMENG (CN)
YU MENG (CN)
LUO MIN (CN)

Application Number:

PCT/CN2019/085625

Publication Date:

January 02, 2020

Filing Date:

May 06, 2019

Export Citation:

Click for automatic bibliography generation Help

Assignee:

TENCENT TECH SHENZHEN CO LTD (CN)

International Classes:

G10L15/26; G10L15/16; G10L21/0216

Foreign References:

CN109272989A	2019-01-25
CN105765650A	2016-07-13
CN104936091A	2015-09-23
JPH04273298A	1992-09-29

Other References: