计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (3): 231-236.DOI: 10.3778/j.issn.1002-8331.1603-0057

• 工程与应用 • 上一篇    下一篇

Android平台说话人认证系统的并行计算与设计

徐利敏1,魏  翔2   

  1. 1.南京财经大学 国际经贸学院 电子商务重点实验室,南京 210046
    2.东南大学 计算机工程与科学学院,南京 211189
  • 出版日期:2017-02-01 发布日期:2017-05-11

Analysis and design of speaker authentication system based on Android platform of parallel computation

XU Limin1, WEI Xiang2   

  1. 1.Key Laboratory of Electronic Business, School of International Economics and Business, Nanjing University of Finance and Economics, Nanjing 210046, China
    2.School of Computer Science and Engineering, Southeast University, Nanjing 211189, China
  • Online:2017-02-01 Published:2017-05-11

摘要: 智能手机技术的快速进步催生了移动商务的蓬勃发展,智能手机移动商务平台的安全性和身份认证问题已经成为移动商务能否进一步发展的关键。说话人识别技术作为一种生物识别认证技术应用到现有的智能手机中非常便利,而且有其他安全认证技术所无法比拟的优势。因此,将说话人识别相关技术应用于Android系统的安全认证中,设计了基于Android平台的说话人认证系统。同时由于智能手机多核性和特征参数提取工作的独立性,为了提高认证系统的效率,提出了基于Android平台的并行算法,并在不同机型上做了相关的实验,通过实验结果可以发现在Android平台采用并行算法能够较大幅度地提高认证系统的效率,从而提高认证系统的用户体验。

关键词: 说话人识别, 认证系统, 并行计算, Android平台, 梅尔倒谱系数

Abstract:  Rapid advances in smartphone technology spawn a booming mobile commerce, security and authentication issues of mobile commerce based on smartphone platform are also prominent, which has become the key to the further development of mobile commerce. Speaker recognition technology is easy to apply to the existing smart phones, and there are some advantages compared with other cryptographic techniques. Therefore, this paper applies speaker recognition technology in safety and authentication based on Android system, devises a speaker authentication system based on Android platform. Simultaneously, because smart phones are mostly multi-core, and the speaker feature extraction is independent, so in order to improve the efficiency of speaker authentication system, this paper presents a parallel algorithm based on the Android platform. Experiments on the different types of smartphones demonstrate that the Android platform using parallel algorithm can improve the efficiency of the authentication system, thereby improving the user experience of the authentication system.

Key words:  speaker recognition, authentication system, parallel computation, Android platform, Mel Frequency Cepstrum Coefficient(MFCC)