基于自给式呼吸器的语音减噪改进方法  

An Improved Method of Voice Noise Reduction Using Self-contained Breathing Apparatus

作　　者：宋宫琨琨 郭业才[1,2] 

机构地区：[1]南京信息工程大学电子与信息工程学院,南京210044 [2]江苏省大气环境与装备技术协同创新中心,南京210044

出　　处：《科学技术与工程》2015年第16期63-67,共5页Science Technology and Engineering

基　　金：江苏省高校自然科学基金重大资助项目(13KJA510001);高校科研成果产业化推进项目(JHB2012-9);江苏省高校"信息与通信工程"优势学科建设工程资助项目;江苏省六大人才高峰项目(2008026);江苏省普通高校研究生科研创新计划项目(SJZZ_0111)资助

摘　　要：针对传统减噪方法在处理自给式呼吸器通信时语音音质不清晰、噪声干扰严重的问题,提出了一种基于自给式呼吸器的语音减噪改进方法,并进行了实验验证。在该改进方法中,先定义一个新的判决变量为相关函数能量,利用该变量进行语音端点检测,然后根据检测结果,分别对有话段语音信号采取声压差值减噪处理,对无话段噪声信号采用静默处理。实测实验结果表明,对于自给式呼吸器采集到的语音信号,改进方法的准确性与鲁棒性要优于传统语音减噪方法。In view of the problems of fuzziness speech quality and serious noise interference by traditional methods when processing the communication voice, an improved method of voice noise reduction using self-contained breathing apparatus was proposed and demonstrated by the practical experiments. In this proposed method, a new decision variable called correlation function and energy （CE） is defined. By using this decision variable, the endpoint of speech can be detected precisely. According to the results of.detection, the noise reduction of sound pressure difference method is used for voice signals and the silence method is used for noise signals. To the voice signals acquired by the self-contained breathing apparatus, the practical experiments results show that the proposed method can get better accuracy and robustness than traditional voice noise reduction methods.

关 键 词：语音端点检测 语音减噪 自给式呼吸器 

分 类 号：TN911[电子电信—通信与信息系统]