Encryption speech content authentication method based on Hash features

Abstract

The present invention relates to an encryption speech content authentication method based on Hash features. The method comprises the steps of: after original speech is input, performing encryption forthe original speech through Logistic mapping and stream cipher RC4 to generate encrypted speech, performing framing for the encrypted speech, executing integer wavelet transform (IWT) and discrete cosine transform (DCT) for each frame of the encrypted speech, comparing a mean value and a variance of a low-frequency DCT coefficient to calculate Hash features, and employing differential expansion to take the Hash features as watermark embedded into the high-order bit of a detail coefficient of the IWT; and executing inverse IWT transform for an IWT approximation coefficient and the detail coefficient comprising the Hash features to obtain the encrypted speech comprising watermark, and extracting the Hash features and the reconstructed Hash features from the encrypted speech comprising watermark to perform comparison to perform content authentication for the encrypted speech. The encryption speech content authentication method improves the robustness of the speech content authenticationin the cloud computing, can accurately located the tempered speech frame and is wider in the application range in the practical application.

Description

technical field [0001] The invention belongs to the field of encrypted voice content authentication, in particular to a hash feature-based encrypted voice content authentication method. Background technique [0002] Speech signal is an important multimedia signal, which can be applied to scenarios such as military command, judicial department recording evidence and online audio instruction. Since voice files require a lot of storage space, many voice files are stored in the cloud. However, this is not safe because anyone can download, read and tamper with the content of the voice file. Therefore, it is very important to protect the voice file content and judge the integrity of the voice file in cloud computing. [0003] Encryption is one of the most effective methods of voice content protection because it converts raw data into incomprehensible data. For information security and privacy protection, data is usually encrypted before being uploaded and transmitted to the clo...

AI Technical Summary

Problems solved by technology

However, these techniques have not been systematically studied and have not been applied to encrypted speech

Also, there are few content authentication schemes for encrypted voice

[0004] In addition, there are still some limitations in the field of speech encryption, such as the content of encrypted speech is random, the features of the original speech disappear, and most of the conventional methods for extracting features from the original speech cannot be directly applied to the field of speech encryption

Images