An Improved Phase Coding Audio Steganography Algorithm

TL;DR

Addresses the threat of AI-generated audio forgery by proposing an improved Phase Coding scheme that segments audio, computes amplitude $A_i$ and phase $\phi_i$ via $FFT$, and embeds data as $\phi_{data}$ into the mid-frequency phase to produce $\phi'_i$, reconstructed via $IFFT$. The approach delivers a five-step, segmentation-based embedding with direct phase updates, achieving enhanced undetectability and data integrity while reducing computational load. BER analyses show lower error rates across varying message lengths and robustness to clipping, indicating improved reliability over traditional Phase Coding. This work advances secure audio watermarking for authenticity verification, with practical implications for voice-identity verification and secure communications.

Abstract

Advances in AI technology have made voice cloning increasingly accessible, leading to a rise in fraud involving AI-generated audio forgeries. This highlights the need to covertly embed information and verify the authenticity and integrity of audio. Digital Audio Watermarking plays a crucial role in this context. This study presents an improved Phase Coding audio steganography algorithm that segments the audio signal dynamically, embedding data into the mid-frequency phase components. This approach enhances resistance to steganalysis, simplifies computation, and ensures secure audio integrity.

Figures (3)

• Figure 1: Phase Comparison of the traditional Phase Coding algorithm
• Figure 2: Phase Comparison of the improved Phase Coding algorithm
• Figure 3: Bit Error Rate (BER) Comparison of Traditional and Improved Algorithms