Research Article
A Reliable Voice Perceptual Hash Authentication Algorithm
@INPROCEEDINGS{10.1007/978-3-030-89814-4_19, author={Li Li and Yang Li and Zizhen Wang and Xuemei Li and Guozhen Shi}, title={A Reliable Voice Perceptual Hash Authentication Algorithm}, proceedings={Mobile Multimedia Communications. 14th EAI International Conference, Mobimedia 2021, Virtual Event, July 23-25, 2021, Proceedings}, proceedings_a={MOBIMEDIA}, year={2021}, month={11}, keywords={Perceptual hash Non-negative matrix factorization MFCC}, doi={10.1007/978-3-030-89814-4_19} }- Li Li
Yang Li
Zizhen Wang
Xuemei Li
Guozhen Shi
Year: 2021
A Reliable Voice Perceptual Hash Authentication Algorithm
MOBIMEDIA
Springer
DOI: 10.1007/978-3-030-89814-4_19
Abstract
In order to protect the authenticity and integrity of voice, this paper proposes a reliable voice perceptual hash authentication algorithm. This algorithm considers the dynamic characteristics of voice signals and proposes a scheme to construct a perceptual feature matrix containing voice static and dynamic characteristics based on the Mel frequency inverted spectrum coefficient and its first- and second-order difference parameters. In the process of perceptual hash chain construction, the feature matrix is degraded by two-dimensional non-negative matrix factorization method, and the result of decomposition is quantified to construct the perceptual hash chain. In the process of voice authentication, hamming distance method is used to measure the distance between perceptual hash chains. Experiments show that the algorithm has good distinguishability and robustness, and the voice perceptual hash authentication can be carried out accurately and reliably.
