Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SPEECH RECOGNITION METHOD, APPARATUS, COMPUTING DEVICE AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/051564
Kind Code:
A1
Abstract:
A speech recognition method, an apparatus, a computing device and a computer non-volatile readable storage medium, which relate to the field of natural language processing. The method comprises: inputting audio data into a speech recognition model to obtain an outputted preliminary recognition result (210); determining an audio segment in audio data corresponding to a text unit in the preliminary recognition result (220); for each text unit, if the sound pressure of the audio segment corresponding to the text unit meets a predetermined condition, replacing the text unit in the preliminary recognition result with a placeholder so as to obtain an intermediate result (230); determining the degree of matching between the intermediate result and each standard text sequence in a preset standard text library (240); and on the basis of the degree of matching, selecting a target standard text sequence from among the standard text sequences in the standard text library as a final recognition result (250). The described method reduces the possibility of speech recognition being inaccurate due to the existence of a text unit for which the sound pressure of a corresponding audio segment does not meet a certain requirement, thereby improving the accuracy of speech recognition.

Inventors:
WANG JIANZONG (CN)
PENG JUNQING (CN)
QU XIAOYANG (CN)
Application Number:
PCT/CN2019/117675
Publication Date:
March 25, 2021
Filing Date:
November 12, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G10L15/22
Foreign References:
CN109920414A2019-06-21
JP2017016131A2017-01-19
CN104952219A2015-09-30
CN108428446A2018-08-21
Attorney, Agent or Firm:
SHENZHEN LUNGTIN LIANDING INTELLECTUAL PROPERTY AGENT LTD. (CN)
Download PDF: