摘 要:MATLAB GUI 是 MATLAB 用户用于算法开发、数据可视化的交互式环境,通过 GUI 生成的操作界面可以使用户不用浏览烦冗的代码而进行操作。文章基于数字信号处理和 MATLAB,运用图形可视化工具 GUI,完成了从倒谱分析出发的数字语音篡改检测界面平台,进行语音的录制,研究了在拼接不同采样率的音频下时域波形、频域波形、倒谱图和声道冲激响应频谱的变化,更直观地分析待检测的语音信号的篡改情况。
关键词:MATLAB GUI;语音篡改;音频特征
DOI:10.19850/j.cnki.2096-4706.2022.17.022
中图分类号:TP391 文献标识码:A 文章编号:2096-4706(2022)17-0085-06
Implementation of Voice Tampering Detection System Based on MFCC
TANG Yijing, ZHOU Chengcheng, WANG Peng, SUN Shifan, WU Wenqi
(School of Information and Communication Engineering, Nanjing Institute of Technology, Nanjing 211167, China)
Abstract: MATLAB GUI is an interactive environment used by MATLAB users for algorithm development and data visualization. The operation interface generated by GUI enables users to operate without browsing the complicated code. Based on digital signal processing and MATLAB, this paper completes the digital voice tampering detection interface platform from cepstrum analysis by using GUI which is a graphic visualization tool. It records the voice, and studies the changes of time domain waveform, frequency domain waveform, cepstrum map and channel impulse response spectrum when splicing different sampling rates of audio, so as to more intuitively analyze the tampering situation of the voice signal to be detected.
Keywords: MATLAB GUI; voice tampering; audio feature
参考文献:
[1] 何钢 . 基于小波神经网络的声景质量预测模型 [D]. 哈尔滨:哈尔滨工业大学,2012.
[2] MANNEPALLI K P,KRISHNA P V,KRISHNA K V,et al.Copy and Move Detection in Audio Recordings Using Dynamic Time Warping Algorithm [J].International Journal of Innovative Technology and Exploring Engineering,2019,9(2):2244-2249.
[3] YAN Q,YANG R,HUANG J W.Robust Copy-Move Detection of Speech Recording Using Similarities of Pitch and Formant [J].IEEE Transactions on Information Forensics and Security,2019,14(9):2331-2341.
[4] 倪令格,王华朋,刘元周,等 . 翻录对语音真实性检验的影响研究 [J].中国人民公安大学学报(自然科学版),2020,26(4):8-14.
[5] 高华,包永强 . 基于时域和频域的语音篡改检测算法[C]//2014’中国西部声学学术交流会论文集,贵阳:《声学技术》编辑部,2014:123-126.
[6] 包永强,梁瑞宇,丛韫,等 . 音频取证若干关键技术研究进展 [J]. 数据采集与处理,2016,31(2):252-259.
[7] 王彪 . 一种基于小波包的语音信号特征提取方法研究 [J].信息技术,2012,36(6):158-160+163.
作者简介:唐依静(2001—),女,汉族,江苏苏州人,本科在读,研究方向:通信工程,信号处理等。