Hide And Speak Towards Deep Neural Networks For Speech Steganography

3 minute read

Published:

Hide and Speak: Towards Deep Neural Networks for Speech Steganography

Felix Kreuk*, Yossi Adi*, Bhiksha Raj, Rita Singh, Joseph Keshet

Below are some samples from our steganography system for you to listen. "Original Carrier" refers to the original audio in which we conceal a hidden message. "Embedded Carrier" refers to the audio with a hidden message, "Original Message" referes to the original message we wish to hide inside "Original Carrier" and "Decoded Message" is the message decoded from "Embedded Carrier".
Note, "Original Carrier" and "Embedded Carrier" should sound identical so as to not raise suspicion."Decoded Message" should be intelligble for a human listener but is allowed to suffer some degradation in quality (espicially for high compression rates).

Paper: https://arxiv.org/abs/1902.03083
Source code: https://github.com/felixkreuk/HideAndSpeak

Single Message

Original CarrierEmbedded CarrierOriginal MessageDecoded Message















3 Messages (x3 compression)

Original CarrierEmbedded CarrierOriginal MessageDecoded Message
















5 Messages (x5 compression)

Original CarrierEmbedded CarrierOriginal MessageDecoded Message