Desynchronization Attacks Resilient Watermarking Method Based on Frequency Singular Value Coefficient Modification

Zhao, Juan; Zong, Tianrui; Xiang, Yong; Gao, Longxiang; Zhou, Wanlei; Beliakov, Gleb

doi:10.1109/taslp.2021.3092555

Cited by 19 publications

(9 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, we embed the original watermark (wm1) using our proposed method. To further embed the attacker's watermark (wm2), we first adopt existing audio watermarking schemes FSVC [23] and RFDLM [21]. As shown in Table V, the original watermark can still be extracted as the evidence of voice cloning attacks, while the attacker fails to extract his watermark.…”

Section: E Resistance Against Adaptive Attacksmentioning

confidence: 99%

“…For robustness, some research has demonstrated resilience against common and unavoidable signal processing distortions [47]. More recently, the focus has shifted toward exploring robustness under intricate conditions, such as rerecording [21], [24], desynchronization distortion [23], [20], cropping distortion [48], etc. In a nutshell, existing audio watermarking can already well balance the trade-off between fidelity and normal robustness.…”

Section: Audio Watermarkingmentioning

confidence: 99%

“…In Sec. VI-A we evaluate two audio watermarking methods (i.e., FSVC [23] and RFDLM [21]), and show that they totally fail to detect voice cloning attacks. Concurrently, deep learning-based end-to-end audio watermarking has recently emerged [24], [25], [49], yet it encounters considerable limitations.…”

Section: Audio Watermarkingmentioning

confidence: 99%

See 2 more Smart Citations

Detecting Voice Cloning Attacks via Timbre Watermarking

Liu,

Zhang,

Zhang

et al. 2024

Proceedings 2024 Network and Distributed System Security Symposium

View full text Add to dashboard Cite

Nowadays, it is common to release audio content to the public, for social sharing or commercial purposes. However, with the rise of voice cloning technology, attackers have the potential to easily impersonate a specific person by utilizing his publicly released audio without any permission. Therefore, it becomes significant to detect any potential misuse of the released audio content and protect its timbre from being impersonated.

show abstract

Section: E Resistance Against Adaptive Attacksmentioning

confidence: 99%

Section: Audio Watermarkingmentioning

confidence: 99%

See 1 more Smart Citation

Detecting Voice Cloning Attacks via Timbre Watermarking

Liu,

Zhang,

Zhang

et al. 2024

Proceedings 2024 Network and Distributed System Security Symposium

View full text Add to dashboard Cite

show abstract

“…Transform domain methods are the most popular due to their capability of exploiting signal characteristics and/or human auditory properties. Typical transform methods include discrete cosine transform (DCT) [2][3][4][5][6], discrete Fourier transform (DFT) [7][8][9][10], discrete wavelet transform (DWT) [3,[11][12][13][14], and singular value decomposition (SVD) [15][16][17]. The above transform-domain methods provide good performance in terms of robustness and imperceptibility; however, the computational overhead is usually higher than that of methods implemented in the time domain.…”

Section: Introductionmentioning

confidence: 99%

FFT-Based Dual-Mode Blind Watermarking for Hiding Binary Logos and Color Images in Audio

Lee

2023

IEEE Access

View full text Add to dashboard Cite

Watermarking technology is a feasible approach to prevent copyright infringement. The practicality of the watermark itself affects the acceptance and penetration of relevant applications. In this study, a dual-mode FFT-based audio watermarking framework is devised to achieve both binary and numeric data hiding concurrently using respective modulations. We demonstrate: (1) the adaptive magnitude modulation (AMM) applied to large-magnitude FFT coefficients is competent to deliver efficient and effective binary watermarking at a rate of 678.46 bits per second; (2) the phase modulation (PM) in charge of the embedding of pixel values directly into medium-size FFT coefficients can operate at a rate of 508.85 numbers per second. With a tactical selection of embedding positions and strengths, both the FFT-based AMM and PM exhibit exceptional robustness against commonly encountered attacks. According to the perceptual evaluation of audio quality (PEAQ) metric, the resulting audio quality remains indistinguishable from the original audio, even at the aforementioned high payload capacity. INDEX TERMSDual-mode blind audio watermarking; Adaptive magnitude modulation; Phase modulation; Color image watermark.

show abstract

“…The rapid development of Internet technology has made multimedia data vulnerable to malicious manipulation and unauthorized distribution, resulting in copyright infringement [1]. Digital watermarking is a promising solution, but there is a trade-off between its imperceptibility and robustness [2].…”

Section: Introductionmentioning

confidence: 99%

A zero‐watermarking technique based on i‐vector model for audio copyright protection

Xu,

He,

Guo

2023

Electronics Letters

View full text Add to dashboard Cite

Audio zero‐watermarking is a promising technology for audio copyright protection. It does not modify the content of the original carrier and has good imperceptibility. Considering that i‐vector model can map audio signals of different lengths into a vector space of fixed length, which has good robustness against desynchronization attacks, a robust audio zero‐watermarking technique based on i‐vector model is proposed here. The proposed scheme extracts the i‐vector of the input signal to identify its inherent stability features and then generates a key by XOR operation on the watermark and encoded feature sequences. Experimental results show that this scheme can effectively resist common attacks and desynchronization attacks.

show abstract

Desynchronization Attacks Resilient Watermarking Method Based on Frequency Singular Value Coefficient Modification

Cited by 19 publications

References 35 publications

Detecting Voice Cloning Attacks via Timbre Watermarking

Detecting Voice Cloning Attacks via Timbre Watermarking

FFT-Based Dual-Mode Blind Watermarking for Hiding Binary Logos and Color Images in Audio

A zero‐watermarking technique based on i‐vector model for audio copyright protection

Contact Info

Product

Resources

About