基于语音增强的远场说话人识别技术  

Overview on the Far Field Speaker Recognition Technology Based on Speech Enhancement

作　　者：覃晓逸 蔡丹蔚 胡伟湘 苗磊 李明 QIN Xiaoyi;CAI Danwei;HU Weixiang;MIAO Lei;LI Ming(School of Electronics and Information Technology,Sun Yat - sen University,Guangzhou,410006,China;Data Science Research Center,Duke Kunshan University,Kunshan,215300,China;Huawei Digital Technologies Co.,Ltd,Beijing,100085,China)

机构地区：[1]中山大学电子与信息工程学院,广州410006 [2]昆山杜克大学大数据研究中心,苏州215300 [3]华为北京技术研究所,北京100085

出　　处：《网络新媒体技术》2019年第4期1-10,共10页Network New Media Technology

摘　　要：随着声纹技术的发展并逐渐应用于智能家居领域,尤其是智能音箱已经开始影响着人们的生活,远场环境下的说话人识别技术提升迫在眉睫。本文针对说话人识别技术的发展现状以及远场环境下语音信号的处理,包括麦克风阵列技术的应用,抗噪、抗混响技术的发展与应用等两个主要方面进行阐述。With the development of speaker recognition technology and gradually applied in the field of smart home,especially with the smart speakers,speaker recognition in the far - field condition becomes more and more important. This paper will focus on the current development of speaker recognition technology,speech signal processing in the far - field environment,including the application of mi- crophone array technology,the development and application of denoise and dereverberation technology.

关 键 词：说话人识别 语音增强 麦克风阵列 远场环境 

分 类 号：TN912.35[电子电信—通信与信息系统]