一种基于线性预测的语音认证算法  

A Speech Authentication Algorithm Based on Linear Prediction Analysis

作　　者：夏西泉[1] 王平[2] 

机构地区：[1]重庆电子工程职业学院应用电子学院,重庆401331 [2]重庆大学输配电装备及系统安全与新技术国家重点实验室,重庆400044

出　　处：《火力与指挥控制》2015年第10期72-74,79,共4页Fire Control & Command Control

基　　金：重庆市科学技术研究基金资助项目(KJ132206)

摘　　要：针对传统语音认证算法不适用于资源受限的移动通信终端语音通信的实时性要求,提出了一种高效的基于线性预测分析的语音认证算法。该算法基于语音信号的线性预测分析的原理对LPC系数参数进行优化,并对构成的LPC系数矩阵进行后处理、非负矩阵分解和量化,形成感知特征序列。实验结果表明:所提算法对于内容保持操作具有良好的鲁棒性,并且运算效率没有降低,可以满足语音通信实时性的要求。According to the situation that traditional speech authentication algorithms does not appropriated for present speech communication,a speech authentication algorithm based on linear prediction analysis is proposed,it can satisfy the requirement of the efficiency and the robustness for speech authentication. Firstly,the LPC coefficients are optimized based on the principle of linear prediction analysis for speech signal,and make Non-negative Matrix Factorization（NMF） toward the coefficients. Finally,the formed getting Characteristics sequences is quantified. Experiments show that the proposed algorithm has good robustness for content preserving operations,and it doesn＇t reduce the efficiency while meeting the robustness,it can satisfy the real-time requirement of speech communication.

关 键 词：语音认证 鲁棒性 线性预测系数 非负矩阵分解 

分 类 号：TN911.72[电子电信—通信与信息系统]