用简易实时通信软件进行的远程面试有替考漏洞. 为此, 本文提出了一种基于语音对比的简便远程身份认证方案: 接受面试者只需要使用通用的通信软件, 不需要安装特殊的软件或特殊硬件系统, 这对于接受面试者在边远地区或国外尤其便利. 主考官在电脑上安装音轨抓取和录屏截屏软件, 当面试者被录取后, 采集现场无损音视频资料, 将此资料和远程获取的同一面试者的音视频资料进行人工比对, 实现以声纹认证为主的身份认证. 为了验证方案的可行性, 本研究进行了两轮实验, 采集了来自7个不同国家的远程语音数据. 通过一系列软件分析和人工比对, 实验结果表明, 本方案身份认证准确率较高, 为后期的全计算机认证打下了基础.
There are vulnerabilities in conducting remote interviews via simple real-time communication software. To this end, this paper proposes a simple and easy remote identity authentication scheme, whereby the recruitee only needs to use general communication software. There is no need to install special software or special hardware systems, which is particularly convenient for recruitees residing in remote areas or international locations. Recruiters need to install audio capture and screen capture software on their computer; once the recruitee is admitted into the interview, the recruiter can achieve identity authentication using audio and video comparison technology. In order to verify the feasibility of this scheme, we carried out two rounds of experiments, collecting remote voice data from seven different countries through a series of software analysis tools as well as manual comparison. The experimental results show that the scheme has a high degree of accuracy for identity authentication, laying the foundation for prospective fully automatic authentication.
[1] 田旭. 基于WebRTC技术的远程面试系统的设计与实现 [D]. 武汉: 华中师范大学, 2014.
[2] 潘迪, 梁士利, 魏莹, 等. 语谱图二次傅里叶变换特定人二字汉语词汇识别 [J]. 东北师大学报(自然科学版), 2017, 49(2): 95-100
[3] DIFFIE W, HELLMAN M E. New directions in cryptography [J]. IEEE Transactions on Information Theory, 1976, 22(6): 644-654
[4] 张俊松, 张启坤, 甘勇, 等. 适用于无线医疗传感网的身份认证协议 [J]. 北京邮电大学学报, 2018, 41(4): 104-109
[5] WOJTOWICZ W, OGIELA M R. Biometric watermarks based on face recognition methods for authentication of digital images [J]. Security and Communication Networks, 2015, 8(9): 1672-1687
[6] WANG C, LI Y, SONG X. Video-to-video face authentication system robust to pose variations [J]. Expert Systems with Applications An International Journal, 2013, 40(2): 722-735
[7] 董博生. 人脸识别技术的实现及其在远程身份验证中的应用 [D]. 北京: 北方工业大学, 2007.
[8] PLAMONDONA R, LORETTEB G. Automatic signature verification and writer identification: The state of the art [J]. Pattern Recognition, 1989, 22(2): 107-131
[9] 曾斌, 姚路, 陈志诚. 基于声纹识别的Web身份认证系统设计 [J]. 计算机工程, 2011, 37(15): 149-151
[10] 科大讯飞股份有限公司. 声纹识别 [EB/OL]. [2020-07-17]. http://www.xfyun.cn/services/isv.
[11] 吴宗济, 林茂灿. 实验语音学概要 [M]. 北京: 高等教育出版社, 1989.
[12] 钟彩顺. 基于元音共振峰的跨语言司法语音比对研究 [J]. 上海外国语大学学报, 2019, 42(1): 61-71
[13] LADEFOGED P, JOHNSON K. A Course in Phonetics [M]. 7th ed. 北京: 北京大学出版社, 2015.
[14] 汪勇, 熊前兴. MP3格式解析 [J]. 计算机应用与软件, 2004, 21(12): 126-128
[15] 余玲飞, 刘强. 基于深度循环网络的声纹识别方法研究及应用 [J]. 计算机应用研究, 2019, 36(1): 153-158
[16] 闫河, 董莺艳, 王鹏, 等. 基于CNN-LSTM网络的声纹识别研究 [J]. 计算机应用与软件, 2019, 36(4): 166-170