Unlock instant, AI-driven research and patent intelligence for your innovation.

Voice recognition method and device and medium

A speech recognition and to-be-recognized technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of inability to customize and low accuracy of speech recognition, and achieve the goal of improving the accuracy of speech recognition, increasing coverage, and improving scores. Effect

Active Publication Date: 2020-05-08
BEIJING SOGOU TECHNOLOGY DEVELOPMENT CO LTD
View PDF13 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The training of the customized model needs to obtain a large amount of relevant corpus in advance for the application scenario. However, for some conferences and other scenarios, because the specific content of the speeches of the participants cannot be obtained in advance, the customization cannot be carried out, which in turn makes the accuracy of speech recognition in specific application scenarios lower. Low

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Voice recognition method and device and medium
  • Voice recognition method and device and medium
  • Voice recognition method and device and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] refer to figure 2 , which shows a flow chart of the steps of Embodiment 1 of a speech recognition method of the present invention, which may specifically include the following steps:

[0067] Step 201, receiving voice data to be recognized, and determining a first keyword related to the voice data to be recognized;

[0068] Step 202, according to the knowledge map, determine the second keyword related to the above-mentioned first keyword;

[0069] Step 203: Decode the speech data to be recognized, adjust the score of the decoding path corresponding to the subsequent speech data to be recognized according to the first keyword and the second keyword, and adjust the score according to the adjusted decoding path score to determine the speech recognition result corresponding to the speech data to be recognized.

[0070] figure 1 At least one step included in the shown method can be executed by the client and / or the server. It can be understood that the embodiment of the ...

Embodiment 2

[0111] refer to Figure 4 , which shows a flow chart of the steps of Embodiment 2 of a speech recognition method of the present invention, which may specifically include the following steps:

[0112] Step 401, receiving voice data to be recognized, and determining a first keyword related to the voice data to be recognized;

[0113] Step 402, according to the knowledge map, determine the second keyword related to the above-mentioned first keyword;

[0114] Step 403, according to the search engine, determine the third keyword related to the above-mentioned first keyword;

[0115] Step 404: Decode the speech data to be recognized, adjust the score of the decoding path corresponding to the subsequent speech data to be recognized according to the first keyword, the second keyword and the third keyword, and adjust the score according to the above-mentioned The adjusted score of the decoding path determines the speech recognition result corresponding to the speech data to be recogn...

Embodiment 3

[0124] refer to Figure 5 , which shows a flow chart of the steps of Embodiment 3 of a speech recognition method of the present invention, which may specifically include the following steps:

[0125] Step 501, receiving speech data to be recognized, and determining a first keyword related to the speech data to be recognized;

[0126] Step 502, according to the knowledge map, determine the second keyword related to the above-mentioned first keyword;

[0127] Step 503: Decode the voice data to be recognized, adjust the score of the decoding path corresponding to the subsequent voice data to be recognized according to the first keyword and the second keyword, and adjust the scores according to the adjusted decoding path Scoring, determining the speech recognition result corresponding to the speech data to be recognized;

[0128] compared to figure 2 In the first embodiment of the method shown, the method of this embodiment may also include:

[0129] Step 504, presenting the ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention provides a voice recognition method and device and a device for voice recognition, and the method specifically comprises the steps: receiving to-be-recognized voice data, and determining a first keyword related to the to-be-recognized voice data; determining a second keyword related to the first keyword according to a knowledge graph; and decoding the to-be-recognized voice data, adjusting the score of a decoding path corresponding to the to-be-recognized voice data according to the first keyword and the second keyword, and determining a voice recognition resultcorresponding to the to-be-recognized voice data according to the adjusted score of the decoding path. According to the embodiment of the invention, the voice recognition accuracy corresponding to the keyword related to the application scene can be improved.

Description

technical field [0001] The present invention relates to the technical field of communication, in particular to a voice recognition method and device, and a machine-readable medium. Background technique [0002] Speech recognition technology can convert speech into corresponding characters or codes, and is widely used in smart home, real-time speech transcription, machine simultaneous interpretation and other fields. Among them, machine simultaneous interpretation is limited by speech recognition technology and machine translation technology, and the quality of machine translation depends on the quality of speech recognition text; therefore, in order to improve the accuracy of machine simultaneous interpretation, it is necessary to improve the quality of speech recognition system. In some application scenarios of machine simultaneous interpretation, problems of recognition and translation of names of people, places, products, or proper nouns are often encountered, and these w...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/26
CPCG10L15/26
Inventor 陈小敏张晶晶陈伟赵超王小川
Owner BEIJING SOGOU TECHNOLOGY DEVELOPMENT CO LTD