说话人辨认系统的研究与实现  

Research and Implementation of Speaker Recognition System

作　　者：陈奕成 殳国华[1] 匡政睿 余玟铮 沈泽宇 Chen Yicheng;Shu Guohua;Kuang Zhengrui;Yu Wenzheng;Shen Zeyu(College of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China)

机构地区：[1]上海交通大学电子信息与电气工程学院,上海200240

出　　处：《电气自动化》2021年第6期83-85,89,共4页Electrical Automation

摘　　要：随着生物特征识别技术逐渐兴起,声纹作为一种具有唯一性、获取方便等特点的生物特征,在身份认证领域中具有广阔的应用前景和较高的市场价值。基于与文本无关的闭集的说话人语音集合,使用Python在Windows操作系统上搭建了一个利用高斯混合模型识别梅尔倒谱系数的具有降噪功能的说话人辨认系统,包括语音信号预处理、声纹特征提取、声纹模型建立与匹配等过程。系统具有识别速度快、识别精度高的特点,为未来的生物特征识别技术提供了新思路。With the gradual development of biometric recognition technology,voiceprint,as a unique and convenient biometric feature,has broad application prospects and high market value in the field of identity authentication.Based on the text-independent closed set speaker speech set,a speaker recognition system with noise reduction function was built on Windows operating system using python based on Gaussian mixture model to recognize mel-frequency cepstrum coefficient(MFCC),including speech signal preprocessing,voiceprint feature extraction,voiceprint model establishment and matching.The system has the characteristics of fast recognition speed and high recognition accuracy,which provides new ideas for future biometric recognition technology.

关 键 词：声纹 文本无关 闭集 高斯混合模型 梅尔倒谱系数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]