摘要
语音增强是语音识别的重要组成部分,它是将尽可能纯净的原始语音从嘈杂的语音中提取出来。其目的是提高语音质量,从而提高语音识别的准确率。笔者在总结近年来的语音增强的相关算法的基础上,将这些算法进行分类,分为基于数字信号处理的方法和机器学习的方法。本文介绍了基于数字信号处理和基于机器学习的语音增强算法,并总结了几种常用语音增强算法的优缺点,最后提出了语音增强算法进一步研究方向。
Speech enhancement is the extraction of original speech as pure as possible from noisy speech,which is an important part of speech recognition.Its purpose is to improve the quality of speech,so as to improve the accuracy of speech recognition.On the basis of summarizing the related algorithms of speech enhancement in recent years,the author classifies these algorithms into digital signal processing methods and machine learning methods.This paper introduces the speech enhancement algorithms based on digital signal processing and machine learning,summarizes the advantages and disadvantages of several commonly used speech enhancement algorithms,and finally puts forward the existing problems of speech enhancement algorithms and further research directions.
作者
曹丽静
CAO Li-jing(School of Information Technology,Hebei University of Economics and Business,Shijiazhuang Hebei 050061,China)
出处
《河北省科学院学报》
CAS
2020年第2期30-36,共7页
Journal of The Hebei Academy of Sciences
关键词
语音增强
语音识别
机器学习
Speech enhancement
Speech recognition
Machine learning
