首页 | 本学科首页   官方微博 | 高级检索  
     

结合隐藏技术的新颖语音检索算法
引用本文:袁旭海,王让定. 结合隐藏技术的新颖语音检索算法[J]. 宁波大学学报(理工版), 2006, 19(4): 435-440
作者姓名:袁旭海  王让定
作者单位:宁波大学,纵横智能软件研究所,浙江,宁波,315211;宁波大学,纵横智能软件研究所,浙江,宁波,315211
基金项目:浙江省自然科学基金;浙江省宁波市博士学科点专项科研基金
摘    要:采用一种新颖的结合信息隐藏技术的语音检索算法,该算法将保密语音和伪语音分割成多段,利用经典的LSB算法分别隐藏到不同的音频,并存储在磁盘空间中.提取语音码流时,利用小波变换第一级细节分量系数能量的大小找到相应的载体音频,提取其中的语音码流,并建立语音库.利用特定人语音检索技术,检出相应的语音码流,合并语音码流得到真正的保密语音.研究表明:新算法的查准率和查全率可以达到很高的水平,为信息隐藏技术的发展开辟了一条新的研究道路,同样扩大了语音检索技术的应用范围。

关 键 词:语音检索  信息隐藏  小波变换  最不重要位
文章编号:1001-5132(2006)04-0435-06
收稿时间:2006-05-30
修稿时间:2006-05-30

A Novel Speech Retrieval Algorithm Combining Information Data Hiding
YUAN Xu-hai,WANG Rang-ding. A Novel Speech Retrieval Algorithm Combining Information Data Hiding[J]. Journal of Ningbo University(Natural Science and Engineering Edition), 2006, 19(4): 435-440
Authors:YUAN Xu-hai  WANG Rang-ding
Affiliation:CKC Software Laboratory, Ningbo University, Ningbo 315211, China
Abstract:In this paper,a novel speech retrieval combining information data hiding technique is presented.Secret speech and fake speech are cut into several parts,and typical LSB(Least Significant Bit) algorithm is used for hiding secret speech and fake speech into different carrier audio while carrier audio is stored in disk space.When speech stream distilled,energy of the first detail level of DWT domain is used for finding corresponding carrier audio.Speech stream is distilled in carrier audio,and speech database is established.Speech retrieval technology is used for finding corresponding speech stream and secret speech is attained by combining speech stream.Experimental result shows that recall ratio and precision ratio can attain a higher level by which a new way is exploited for information hiding technology,and application field of speech retrieval is enlarged.
Keywords:speech retrieval   information hiding   wavelet transform   least significant bit
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号