语音信号特征参数的分析和选取  

Analysis and Selection of Speech Signal Feature Parameters

作　　者：李亭亭[1] 邓丽[1] 许博[1] 段从武 Li Tingting;Deng Li;Xu Bo;Duan Congwu(Sichuan University,Chengdu Sichuan 610065,China)

机构地区：[1]四川大学,四川成都610065

出　　处：《信息与电脑》2018年第5期45-46,49,共3页Information & Computer

摘　　要：在与文本有关的说话人识别系统中,既需要识别说话人的身份,又需要识别语音文本的内容。语音信号特征参数的选取对系统来说至关重要。目前,在传统语音识别系统的研究中,主要采用MFCC参数作为特征参数进行识别。笔者对语音信号特征参数进行分析,对不同的语音特征参数组合进行实验。实验结果证明,在该系统中,MFCC参数与基音参数的组合提高了系统的识别率。In the study of text-related speaker recognition system,it is to include the identity recognition as well as the speech text recognition.The selection of speech signal feature parameters is very important to the system.At present,in the research of traditional speech recognition system,MFCC parameters are used as feature parameters to identify.In this paper,the characteristic parameters of speech signals are analyzed,and the combination of different speech feature parameters is studied.The experimental results show that in this system,the combination of MFCC parameters and pitch parameters improves the recognition rate of the system.

关 键 词：说话人识别技术 语音特征参数 MFCC 基音周期 

分 类 号：TN912.34[电子电信—通信与信息系统]