MLPD: a multi-layer protection with deduplication technique to preserve audio file transmission over the public domain

Sharing confidential voice-based messages over a public network or shared drive is quite challenging. There are some sensitive and highly confidential messages preferred to share through a voice message. The audio cryptography technique is the preferred way to transfer the content over a network. The proposed methodology multi-layer protection with deduplication (MLPD) enhance the security of voice messages by converting voice as an array of numeric values (Matrix) and applying multiple protection layers like (shuffling, transpose, and byte swap) and doing the reverse process in the receiver end with the help of the digital signature metadata file. Meanwhile, sharing the same message over a shared environment is another challenge. The encrypted redundant voice messages may flood the shared environment. The proposed model acknowledges the quality of encryption over voice data. The uniqueness of the proposed work is the performance of deduplication operation over the encrypted messages without any hashing techniques. The results show that MLPD outperforms advanced encryption standard and Fernet algorithms.