A retrieval method for encrypted speech based on improved power normalized cepstrum coefficients and perceptual hashing

doi:10.1007/s11042-022-12560-5

	A retrieval method for encrypted speech based on improved power normalized cepstrum coefficients and perceptual hashing
	Zhang, Qiu-yu; Bai, Jian; Xu, Fu-jiu
刊名	Multimedia Tools and Applications
	2022-05-01
卷号	81 期号:11 页码:15127-15151
关键词	Authentication Chaotic systems Discrete wavelet transforms Efficiency Extraction Hamming distance Hash functions Information retrieval Principal component analysis Speech Cepstrum Chaotic mapping Encrypted speech Encrypted speech retrieval Features extraction Henon chaotic mapping Perceptual hashing Power Power normalized cepstrum coefficient Speech feature extraction Speech features Speech retrieval
ISSN号	1380-7501
DOI	10.1007/s11042-022-12560-5
英文摘要	In order to improve the impact of noise on the robustness and discrimination of the speech perceptual hashing scheme, improve retrieval efficiency and retrieval accuracy, and protect the privacy of the cloud speech data, a retrieval method for encrypted speech based on improved power normalized cepstrum coefficients (PNCC) and perceptual hashing was proposed in the paper. Firstly, the original speech was encrypted by Henon chaotic map inter-frame scrambling encryption algorithm before uploading to the encrypted speech library in cloud server. Secondly, the discrete wavelet transform (DWT) and first-order difference coefficient were used to improve the PNCC feature extraction algorithm to extract speech features, and the principal component analysis (PCA) was used to reduce high-dimensional audio features to one dimension to form frame features that can represent the speech segment. Finally, the frame features are constructed as binary hashing sequences using hash functions and upload it to the system hashing index table in the cloud. When the user retrieves, the hashing sequence of query speech is extracted and matched with the encrypted speech features by normalized hamming distance in the cloud system hashing index table to obtain the retrieval result. Experimental results show that compared with the existing methods, the proposed method has good robustness and discrimination, and improves retrieval efficiency and retrieval accuracy, the security of cloud speech data is improved. In addition, the proposed method has good recognition ability under simulated real noise environment. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.
语种	英语
出版者	Springer
内容类型	期刊论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/158417]
专题	计算机与通信学院
作者单位	School of Computer and Communication, Lanzhou University of Technology, Lanzhou; 730050, China
推荐引用方式 GB/T 7714	Zhang, Qiu-yu,Bai, Jian,Xu, Fu-jiu. A retrieval method for encrypted speech based on improved power normalized cepstrum coefficients and perceptual hashing[J]. Multimedia Tools and Applications,2022,81(11):15127-15151.
APA	Zhang, Qiu-yu,Bai, Jian,&Xu, Fu-jiu.(2022).A retrieval method for encrypted speech based on improved power normalized cepstrum coefficients and perceptual hashing.Multimedia Tools and Applications,81(11),15127-15151.
MLA	Zhang, Qiu-yu,et al."A retrieval method for encrypted speech based on improved power normalized cepstrum coefficients and perceptual hashing".Multimedia Tools and Applications 81.11(2022):15127-15151.