Zixian GE, Yongbing ZHANG, Liang ZHOU, Qiuyu ZHANG

An efficient retrieval algorithm of encrypted speech based on inverse fast Fourier transform and measurement matrix

In this paper, we present an efficient retrieval algorithm for encrypted speech based on an inverse fast Fouriertransform and measurement matrix. Our approach improves query performance, as well as retrieval efficiency andaccuracy, compared to existing content-based encrypted speech retrieval methods. Our proposed algorithm constructsa perceptual hash scheme using perceptual hash sequences from original speech files. By classifying the sequences andapplying run-length compression, we decrease the cloud storage required for the hash index. We secure the speechdatabase by encrypting it with Henon chaos scrambling, which offers excellent resistance to attacks. Experimentalresults show that the robustness, discrimination, and feature extraction efficiency of our proposed method are betterthan the existing alternatives, with good recall and precision ratios and with high retrieval efficiency and accuracy.

PDF

___

[1] Thangavel M, Varalakshmi P, Renganayaki S, Subhapriya GR, Preethi T, Banu AZ. SMCSRC-Secure multimedia content storage and retrieval in cloud. In: IEEE 2016 Recent Trends in Information Technology; 8–9 April 2016; Chennai, India. New York, NY, USA: IEEE. pp. 1-6.
[2] Vavrek J, Viszlay P, Lojka M, Juhar J, Pleva M. Weighted fast sequential DTW for multilingual audio Query-byExample retrieval. J Intell Inf Syst 2018; 2018: 1-17.
[3] Xiao X, Wang JQ. Improved lattice-based speech keyword spotting algorithm. Journal of Tsinghua University 2015; 55: 508-513 (in Chinese).
[4] Zhao W. A high efficient music retrieval algorithm based on content. In: IEEE 2016 Measuring Technology and Mechatronics Automation; 11–12 March 2016; Macau, China. New York, NY, USA: IEEE. pp. 12-15.
[5] Dorfer M, Arzt A, Widmer G. Towards end-to-end audio-sheet-music retrieval. arXiv preprint, arXiv:1612.05070, 2016.
[6] Qin J, Liu X, Lin H. Audio retrieval based on manifold ranking and relevance feedback. Tsinghua Sci Technol 2015; 20: 613-619.
[7] Lotia P, Khan DM. Significance of complementary spectral features for speaker recognition. International Journal of Research in Computer and Communication Technology 2013; 2: 579-588.
[8] Li JF, Wu T, Wang HX. Perceptual hashing based on correlation coefficient of MFCC for speech authentication. Journal of BUPT 2015; 38: 89-93 (in Chinese).
[9] Zhang QY, Xing PF, Huang YB, Dong RH, Yang ZP. Perceptual hashing algorithm for multi-format. Journal of BUPT 2016; 39: 77-82 (in Chinese).
[10] Chen N, Xiao HD, Zhu J. Robust audio fingerprinting based on GammaChirp frequency cepstral coefficients and chroma. Electron Lett 2014; 50: 241-242.
[11] Zhang XZ, Wang YS, Zeng Z, Niu B. An efficient filtering-and-refining retrieval method for big audio data. Journal of Computer Research and Development 2015; 52: 2025-2032 (in Chinese).
[12] Coover B, Han J. A power mask based audio fingerprint. In: IEEE 2014 Acoustics, Speech and Signal Processing; 4–9 May 2014; Florence, Italy. New York, NY, USA: IEEE. pp. 1394-1398.
[13] Stanko T, Chen B, Skoric B. Fingerprint template protection using minutia-pair spectral representations. arXiv preprint, arXiv:1804.01744, 2018.
[14] Patel VM, Ratha NK, Chellappa R. Cancelable biometrics: a review. IEEE Signal Proc Mag 2015; 32: 54-65.
[15] Kaur H, Khanna P. Random distance method for generating unimodal and multimodal cancelable biometric features. IEEE T Inf Foren Sec 2019; 14: 709-719.
[16] Topcu B, Karabat C, Azadmanesh M, Erdogan H. Practical security and privacy attacks against biometric hashing using sparse recovery. EURASIP J Adv Signal Proc 2016; 1: 100-120.
[17] Topcu B, Karabat C, Erdogan H. Unpredictability assessment of biometric hashing under naive and advanced threat conditions. In: IEEE 2016 Signal Processing Conference; 2016; Florence, Italy. New York, NY, USA: IEEE. pp. 2265-2269.
[18] Hine GE, Maiorana E, Campisi P. A zero-leakage fuzzy embedder from the theoretical formulation to real data. IEEE T Inf Foren Sec 2017; 12: 1724-1734.
[19] Wang H, Zhou L, Zhang W, Liu S. Watermarking-based perceptual hashing search over encrypted speech. In: Springer 2013 International Workshop on Digital Watermarking; 1–4 October 2013; Auckland, New Zealand. Berlin, Heidelberg: Springer. pp. 423-434.
[20] Ibrahim A, Jin H, Yassin AA, Zou D. Secure rank-ordered search of multi-keyword trapdoor over encrypted cloud data. In: IEEE 2012 Asia-Pacific Services Computing Conference; 6–8 December 2012; Guilin, China. New York, NY, USA: IEEE. pp. 263-270.
[21] Wang HX, Hao GY. Perceptual speech hashing algorithm based on time and frequency domain change characteristics. China Patent No. 2015102405844, 2015.
[22] Lin L. Study on retrieval for encrypted speech and recovery watermarking-based speech authentication. MSc, Southwest Jiaotong University, Chengdu, China, 2015 (in Chinese).
[23] Zhao H, He S. A retrieval algorithm for encrypted speech based on perceptual hashing. In: IEEE 2016 Natural Computation, Fuzzy Systems and Knowledge Discovery; 13–15 August 2016; Changsha, China. New York, NY, USA: IEEE. pp. 1840-1845.
[24] He SF, Zhao H. A retrieval algorithm of encrypted speech based on syllable-level perceptual hashing. Comput Sci Inf Syst 2017; 14: 703-718.
[25] Glackin C, Chollet G, Dugan N, Cannings N, Wall J, Tahir S, Rajarajan M. Privacy preserving encrypted phonetic search of speech data. In: IEEE 2017 International Conference on Acoustics, Speech and Signal Processing; 5–9 March 2017; New Orleans, LA, USA. New York, NY, USA: IEEE. pp. 6414-6418.
[26] Wang Y, Wang Y, Shi Q. Optimized signal distortion for PAPR reduction of OFDM signals with IFFT/FFT complexity via ADMM approaches. IEEE T Signal Proc 2019; 67: 399-414.
[27] Huang DM, Geng X, Wei LF, Su C. A secure query scheme on encrypted remote sensing images based on Henon mapping. Journal of Software 2017; 27: 1729-1740 (in Chinese).
[28] Wang XW, Cui GW, Wang L, Jia XL, Nie W. Construction of measurement matrix in compressed sensing based on balanced Gold sequence. Chinese Journal of Scientific Instruments 2014; 35: 97-102 (in Chinese).
[29] Zhang WY, Wei ZW, Wang BH, Xiao PH. Measuring mixing patterns in complex networks by Spearman rank correlation coefficient. Physica A 2016; 451: 440-450.
[30] Zhang QY, Hu WJ, Qiao SB. Speech perceptual hashing authentication algorithm based on spectral subtraction and energy to entropy ratio. International Journal of Network Security 2017; 19: 752-760.
[31] Li K, Huang Z, Cheng YC, Lee CH. A maximal figure-of-merit learning approach to maximizing mean average precision with deep neural network based classifiers. In: IEEE 2014 International Conference on Acoustics, Speech and Signal Processing; 4–9 May 2014; Florence, Italy. New York, NY, USA: IEEE. pp. 4503-4507.