Robust Speech Perception Hashing Authentication Algorithm Based on Spectral Subtraction and Multi-feature Tensor

In order to make the speech perception hashing authentication algorithm has strong robustness and discrimination to content preserving operations and speech communication under the common background noise, a new robust speech perceptual hashing authentication algorithm based on spectral subtraction and multi-feature tensor was proposed. The proposed algorithm uses spectral subtraction method to denoise the speech which processed by applying pre-processing. Then, the algorithm acquires each speech component wavelet packet decomposition, MFCC and LPCC feature of each speech component are extracted to constitute the speech feature tensor. The feature tensor is decomposed tensor decomposition to reduce the complexity. Finally, speech authentication is done by generating the hashing values which use mid-value. Experimental results show that the proposed algorithm can denoise the speech effectively, and have good robustness and discrimination to content preserving operations, as well as able to resist the attack of the background noise, which is commonly heard during the communication.