基于Praat软件的电子伪装语音识别研究  

Analysis of Electronic Camouflage Speech Based on Praat Software

作　　者：胡晓光[1] 刘天宇[1] 杨昊易 蔡能斌 HU Xiaoguang;LIU Tianyu;YANG Haoyi;CAI Nengbin(School of Investigation,People's Public Security University of China,Beijing 10038,China;Shanghai Key Laboratory of Crime Scene Evidence,Shanghai 200083,China)

机构地区：[1]中国人民公安大学侦查学院,北京100038 [2]上海市现场物证重点实验室,上海200083

出　　处：《中国人民公安大学学报（自然科学版）》2021年第2期1-6,共6页Journal of People’s Public Security University of China(Science and Technology)

基　　金：上海市现场物证重点实验室开放课题基金(2020XCWZK05);中国人民公安大学专项项目(2020JWCX08);国家重点研究计划项目。

摘　　要：为探究变音软件的变音规律,总结变音软件变音原理。以市面上常见的4款变音软件作为研究对象,利用praat软件对3名实验者变音前后的声音参数(音强、基频、共振峰频率等)进行测定与分析,并对变声结果进行变调分析。发现变音软件主要有改变基频和进行复杂运算两种变音原理,基于此提出对变声进行相应比例的变调,可以得到与原声听觉效果和声学参数上均契合较好的恢复语音。在案件的侦查初期提供指向嫌疑人的线索,也为打击电信诈骗的违法犯罪活动提供了一种新思路。To explore the law of voice changer,and summarize the principle of voice changer.In this paper,four popular voice changer on the market are used as research objects.Praat software is used to measure and analyze the sound parameters(pitch intensity,fundamental frequency,formant frequency,etc.)of three experimenters before and after the tone changing,and the tone changing results are analyzed.It is found that the voice changer mainly has two principles of changing the fundamental frequency and complex arithmetic.Based on this,it puts forward the corresponding proportion of the tone change to the sound change,which can get the sound recovery better in accordance with the original sound auditory effect and acoustic parameters.Providing clues to suspects in the early investigation of the case also provides a new way of thinking for cracking down on illegal and criminal activities of telecom fraud.

关 键 词：praat 电子伪装语音 测定与分析 

分 类 号：D918.2[政治法律—法学]