干涉型光纤语音传感器及语音降噪方法研究

发布时间：2018-02-21 08:44

本文关键词： 光纤传感器语音干涉小波包降噪阈值函数　出处：《仪器仪表学报》2017年11期 　论文类型：期刊论文

【摘要】：近年来,光纤语音传感器已经引起科研界的广泛关注,如何提高语音探测质量是所有类型光纤语音传感器都面临一个关键问题。提出了一种新型小波包语音提取方法,用来提高光纤语音传感器语音探测质量。基于光纤干涉仪原理,搭建了光纤语音传感实验系统。实验结果表明,无论是与硬阈值处理方法还是与软阈值处理方法相比,新阈值处理方法都提高了信噪比。并且保证了语音信号的连续性,改善了语音质量。本方法可广泛应用于语音探测领域,特别是实时语音探测。
[Abstract]:In recent years, optical fiber sensor speech has aroused widespread concern in scientific research, how to improve the quality of speech detection is all types of fiber sensor speech are facing a key problem. A new wavelet packet speech extraction method, optical fiber sensor used to improve speech speech detection quality. Based on the principle of optical fiber interferometer, optical fiber sensor to build voice the experimental system. The experimental results show that both with the soft thresholding method is compared with the hard thresholding method, a new thresholding method can improve the signal-to-noise ratio. And ensure the continuity of the speech signal, improve the speech quality. This method can be widely used in the field of speech detection, especially for real-time speech detection.

【作者单位】：安徽师范大学物理与电子信息学院;光电材料科学与技术安徽省重点实验室;安徽师范大学数学计算机科学学院;
【基金】：安徽省科技攻关项目(1501041137) 安徽省教育厅自然科学基金(KJ2015A079) 安徽师范大学博士科研启动基金(751701)项目资助
【分类号】：TP212

【相似文献】