语音信号声纹识别的研究和仿真  

Research and Simulation of Voice Signal Voiceprint Recognition

作　　者：曾霞[1] 李霜 ZENG Xia;LI Shuang(Guangzhou University Sontan College,Guangdong Guangzhou 511370,China;Shutang Information Technology(Shenzhen)Co.,Ltd.,Guangdong Shenzhen 518000,China)

机构地区：[1]广州大学松田学院,广东广州511370 [2]舒糖讯息科技(深圳)有限公司,广东深圳518000

出　　处：《新一代信息技术》2019年第16期61-65,共5页New Generation of Information Technology

基　　金：2018年广州大学松田学院科研规划项目“语音信号处理及识别应用”(项目编号:Gzdxstxy2018-06)。

摘　　要：随着人工智能技术的日益发展,语音识别的研究和应用得到越来越多的重视,本文就语音信号声纹识别系统进行研究和仿真,通过对语音信号的预处理,提取MFCC特征参数,根据参数进行DTW的识别,从而识别出说话人,实现声纹识别,实验仿真效果表明算法识别效果较好。With the development of artificial intelligence technology,the research and application of speech recognition has received more and more attention.This paper studies and simulates the voice signal voiceprint recognition system,and extracts the MFCC characteristic parameters by preprocessing the speech signal.The parameters are identified by DTW to identify the speaker and realize voiceprint recognition.The experimental simulation results show that the algorithm has better recognition effect.

关 键 词：MFCC系数 声纹识别 人工智能 

分 类 号：TN957.51[电子电信—信号与信息处理]