普通话水平测试系统中语音识别和语音评测技术研究  

Research on Speech Recognition and Evaluation Technology in Putonghua Proficiency Test System

作　　者：廖俐[1] LIAO Li(Department of Education and Art Design,Guangxi Vocational&Technical Institute of Industry,Nanning,Guangxi Province,531000 China)

机构地区：[1]广西工业职业技术学院,广西南宁531000

出　　处：《科技创新导报》2020年第30期84-87,共4页Science and Technology Innovation Herald

基　　金：广西工业职业技术学院2016年度专项教学改革项目“学前教育专业学生语言专长培养的研究与实践”(项目编号:桂工业院[2017]4号);2017年国家社科基金重点项目“出土战国文献汇释今译暨数据库建设”(17AYY014,张玉金主持)阶段成果。

摘　　要：目前我国的普通话测试系统具备语音识别功能和评测语音内容的基础技术,该系统与不同设备的适配条件进行研究,把智能理念融入对测试系统的设计和应用中,使人们可以独立使用测试系统,逐步完成对自身普通话水平的检测和提升。本文对普通话水平测试系统的用户使用进行了全面分析,通过阐述科研人员应当进行的计算机语言学技术研究,把合成技术的应用和语音数据库的应用做了简单剖析。本文根据语音识别在功能和技术方面的固有需求,对其与用户对接中要着重研究的评测技术,进行了功能考察、用户实验等方面的研究阐述。结合用户水平和智能理念,对可实现的技术条件进行了系统分析,提出了优化语音识别、控制评测技术的分析结论。At present,our countr’s Putonghua test system has the function of speech recognition and the basic technology of evaluating voice content.The system studies the adaptation conditions of different equipment,integrates the intelligent concept into the design and application of the test system,so that people can use the test system independently,and gradually complete the detection and improvement of their own Putonghua level.This paper makes a comprehensive analysis of the use of the Putonghua proficiency test system.By describing the computer linguistics technology research that researchers should carry out,it makes a simple analysis of the application of synthesis technology and voice database.According to the inherent requirements of speech recognition in terms of function and technology,this paper focuses on the evaluation technology in the connection with users,including functional investigation,user experiment,etc.Combined with the concept of user level and intelligence,the realizable technical conditions are analyzed systematically,and the analysis conclusion of optimizing speech recognition and control evaluation technology is put forward.

关 键 词：普通话水平测试 系统 语音识别 语音评测技术 

分 类 号：H102[语言文字—汉语]