Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Audio processing method and device, language model training method and device and computer equipment

A language model and audio processing technology, applied in the field of signal processing, can solve the problem of wrong judgment of test takers, and achieve the effect of reducing misjudgment, improving accuracy, and enhancing the probability of occurrence

Active Publication Date: 2020-11-13
TENCENT TECH (SHENZHEN) CO LTD
View PDF9 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In the above process, because the scoring points of the questions are often controlled by keywords, for example, in the question-and-answer question "How do you go home after school (how do you go home after school)", if the ASR model converts the examinee's audio from "Walking home (walking home)" is mistakenly identified as "Working home (working at home)", then due to the wrong identification of the keyword "Walking", it will lead to wrong judgments on the examinee's performance. Therefore, there is an urgent need for an audio The method of processing the accuracy rate of the process

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
  • Audio processing method and device, language model training method and device and computer equipment
  • Audio processing method and device, language model training method and device and computer equipment
  • Audio processing method and device, language model training method and device and computer equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0065] In order to make the purpose, technical solution and advantages of the present application clearer, the implementation manners of the present application will be further described in detail below in conjunction with the accompanying drawings.

[0066] In this application, the terms "first" and "second" are used to distinguish the same or similar items with basically the same function and function. It should be understood that "first", "second" and "nth" There are no logical or timing dependencies, nor are there restrictions on quantity or order of execution.

[0067] In the present application, the term "at least one" means one or more, and the meaning of "multiple" means two or more, for example, a plurality of first positions means two or more first positions.

[0068] Artificial Intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intellige...

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 invention discloses an audio processing method and device, a language model training method and device and system, computer equipment and a storage medium, and belongs to the technical field of signal processing. A phoneme sequence of a target audio is obtained; a first word sequence is obtained based on a traditional voice recognition mode; in addition, contextual information is introduced tocarry out voice recognition under a specific context again, so a second word sequence conforming to the contextual information of the specific context is obtained; the first word sequence and the second word sequence are comprehensively considered; the final semantic information is planned and decoded, equivalently, by introducing the second word sequence, the occurrence probability of some wordsconforming to the specific context in the semantic information is enhanced, the misjudgment situation of some keywords in the semantic information is reduced, the accuracy of the automatic speech recognition process is improved, and thus the accuracy of the audio processing process is improved.

Description

technical field [0001] The present application relates to the technical field of signal processing, in particular to an audio processing method, a language model training method, a device, a system, a computer device and a storage medium. Background technique [0002] In the field of signal processing, Automatic Speech Recognition (ASR) is a hot topic. As a technology that converts human speech into text, ASR can be applied to the intelligent marking system of oral examinations in the field of education. [0003] In the intelligent marking system, ASR technology can be used to recognize candidates' audio into text, and then perform intelligent marking and scoring based on specified rules (such as keyword matching), machine learning or natural language processing (Natural Language Processing, NLP) and other systems. The current ASR technology uses the candidate's audio as the input signal, extracts the waveform, frequency spectrum, pitch and other characteristics of the candi...

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): G10L15/19G10L15/02G10L15/06
CPCG10L15/02G10L15/063G10L15/19G10L2015/025
Inventor 黄江泉
Owner TENCENT TECH (SHENZHEN) CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com