SPEECH RECOGNITION METHOD, APPARATUS AND DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM

Title:

SPEECH RECOGNITION METHOD, APPARATUS AND DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM

Document Type and Number:

WIPO Patent Application WO/2020/238341

Kind Code:

A1

Abstract:

Provided are a speech recognition method, apparatus and device, and a computer-readable storage medium, belonging to the technical field of artificial intelligence, particularly applicable to human-machine interactions in an automobile or an electric vehicle. The method comprises: acquiring or generating a dynamic target language model according to reply information of a first intention, wherein the dynamic target language model comprises a front-end portion and a core portion (201); acquiring a speech signal, and parsing the speech signal to generate keywords (202); and calling the dynamic target language model to determine a second intention and service content, wherein the front-end portion of the dynamic target language model parses the second intention according to the keywords, and the core portion of the dynamic target language model parses the service content according to the keywords (203). The speech recognition method can realize ignoring of irrelevant speech, recognize discontinuous multi-intention speech, and prevent the provided service content from deviating from a user need, and has a good recognition effect.

Inventors:

NIE WEIRAN (CN)
WENG FULIANG (CN)
HUANG YOUJIA (CN)
YU HAI (CN)
HU SHUMANG (CN)

Application Number:

PCT/CN2020/079522

Publication Date:

December 03, 2020

Filing Date:

March 16, 2020

Export Citation:

Click for automatic bibliography generation Help

Assignee:

HUAWEI TECH CO LTD (CN)

International Classes:

G10L15/18; G10L15/22

Foreign References:

CN106486120A	2017-03-08
CN105632495A	2016-06-01
CN105590626A	2016-05-18
CN101656799A	2010-02-24
CN105529030A	2016-04-27
CN109616108A	2019-04-12
US8990085B2	2015-03-24
CN201910470966A	2019-05-31

Other References: