作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2013, Vol. 39 ›› Issue (5): 277-279. doi: 10.3969/j.issn.1000-3428.2013.05.061

• 开发研究与工程应用 • 上一篇    下一篇

基于随机共振的语音真实性检测方法

潘 平,罗 辉,李换双   

  1. (贵州大学计算机科学与信息学院,贵阳 550025)
  • 收稿日期:2012-03-26 出版日期:2013-05-15 发布日期:2013-05-14
  • 作者简介:潘 平(1962-),男,副教授,主研方向:信息处理,信息安全;罗 辉、李换双,硕士研究生
  • 基金资助:
    国家科技计划基金资助项目(2008RR0003);贵州省国际科技合作计划基金资助项目([2009]700109);贵州省科学技术基金资助项目([2012]2132)

Voice Authenticity Detection Method Based on Stochastic Resonance

PAN Ping, LUO Hui, LI Huan-shuang   

  1. (Institute of Computer Science and Information, Guizhou University, Guiyang 550025, China)
  • Received:2012-03-26 Online:2013-05-15 Published:2013-05-14

摘要: 已有的语音资料真实性检测方法是对内容或背景噪声进行的检测,导致对语音信号真实性检测效果不佳。为此,根据数字存储的本质特征和随机共振理论,提出一种语音信号真实性检测方法,从磁/荷的微弱变化中鉴别出编辑点。经大量的时域仿真实验,并与其他方法相比,证明该方法具有较高的鉴别率,并能判断出语音编辑点的位置。

关键词: 数字存储, 随机共振, 语音, 时域, 真实性

Abstract: For now, some researches on the detection of voice authenticity are almost based on the recording material content and background noise, not involves the essence of recording material storage, resulting in poor authenticity detection results of the voice signal. According to the essence of digital storage and the stochastic resonance theory, a test method for voice signal authenticity is proposed. It can identify the edit points from the faint change of magnetism/charge. Many time domain simulation experiments show that this method has higher identification percentage than other methods, and can judge the location of the voice edit points.

Key words: digital storage, stochastic resonance, voice, time domain, authenticity

中图分类号: