Speech watermarking with objective fidelity and robustness criteria

Gurijala, Aparna; Deller, J.R.

doi:10.1109/acssc.2003.1292314

Cited by 8 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, the problem is reduced to a scalar subproblem consisting in computing (i.e., the residual entropy in one dimension). This result is used in Appendix A, under the assumption of , to show that the residual entropy per dimension is given by (29) where is the th harmonic number, is the embedding distortion according to (26), and we have taken into account that for the cubic lattice .…”

Section: A Exact Computation For the Cubic Latticementioning

confidence: 99%

“…Since we are interested in obtaining the behavior for large , we make use of the approximation , which is asymptotically tight for large , with the harmonic number and the Euler-Mascheroni constant, defined as . In this case we have, using (29) (44) Thus, the variance per dimension approximately decreases with the inverse of the squared number of observations. This bound can even be compared to the exact error variance of the optimal dither estimator in order to check the tightness of the bound.…”

Section: Bounds On the Estimation Errormentioning

confidence: 99%

“…Indeed, the information-theoretic formulation given in Section II is closely related to the theory of set-membership estimation (SME) (also known as set-theoretic estimation) [27], [28], which is widely known in the field of automatic control and in certain signal processing areas, such as image recovery. 9 In the set-membership formulation of a problem with solution space , the th observation is associated with a subset that contains all estimates which are consistent with that observation; formally, can be expressed as (47) 9 Interestingly, the set-membership framework has been previously applied to watermark embedding in speech signals [29].…”

Section: Practical Algorithms For Secret Dither Estimationmentioning

confidence: 99%

See 2 more Smart Citations

Security of Lattice-Based Data Hiding Against the Watermarked-Only Attack

Pérez-Freire

Pérez-González

2008

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Abstract-Security of quantization index modulation (QIM) watermarking methods is usually sought through a pseudorandom dither signal which randomizes the codebook. This dither plays the role of the secret key (i.e., a parameter only shared by the watermarking embedder and decoder), which prevents unauthorized embedding and/or decoding. However, if the same dither signal is reused, the observation of several watermarked signals can provide sufficient information for an attacker to estimate the dither signal. This paper focuses on the cases when the embedded messages are either known or constant. In the first part of this paper, a theoretical security analysis of QIM data hiding measures the information leakage about the secret dither as the mutual information between the dither and the watermarked signals. In the second part, we show how set-membership estimation techniques successfully provide accurate estimates of the dither from observed watermarked signals. The conclusion of this twofold study is that current QIM watermarking schemes have a relative low security level against this scenario because a small number of observed watermarked signals yields a sufficiently accurate estimate of the secret dither. The analysis presented in this paper also serves as the basis for more involved scenarios.

show abstract

Section: A Exact Computation For the Cubic Latticementioning

confidence: 99%

Section: Bounds On the Estimation Errormentioning

confidence: 99%

Section: Practical Algorithms For Secret Dither Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Security of Lattice-Based Data Hiding Against the Watermarked-Only Attack

Pérez-Freire

Pérez-González

2008

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

show abstract

“…3 The watermarked signal (stegosignal) is constructed by employing the perturbed LP coefficients, say , and the exact prediction residual in the FIR filter (12) Parametric watermarking was found to be robust against a wide variety of attacks, as discussed in [6]. In particular, the technique is highly robust to additive white noise for two reasons: First, the solution has been shown to be asymptotically immune to additive white noise [73], and second, the parameter estimation process severely attenuates the noise energy in the parameter domain with respect to the signal domain [74].…”

Section: ) Parameter-embedded Watermarkingmentioning

confidence: 99%

“…Filtering attacks are reported to be more challenging in that account. More recent efforts have concentrated on robustness to filtering with improved results [6], [71], [73]. Because of the correlation in the noise, however, the solution is biased asymptotically.…”

Section: ) Parameter-embedded Watermarkingmentioning

confidence: 99%

SpeechFind: advances in spoken document retrieval for a National Gallery of the Spoken Word

Hansen

Huang

Zhou

et al. 2005

IEEE Trans. Speech Audio Process.

Self Cite

View full text Add to dashboard Cite

Abstract-Advances in formulating spoken document retrieval for a new National Gallery of the Spoken Word (NGSW) are addressed. NGSW is the first large-scale repository of its kind, consisting of speeches, news broadcasts, and recordings from the 20th century. After presenting an overview of the audio stream content of the NGSW, with sample audio files from U.S. Presidents from 1893 to the present, an overall system diagram is proposed with a discussion of critical tasks associated with effective audio information retrieval. These include advanced audio segmentation, speech recognition model adaptation for acoustic background noise and speaker variability, and information retrieval using natural language processing for text query requests that include document and query expansion. For segmentation, a new evaluation criterion entitled fused error score (FES) is proposed, followed by application of the CompSeg segmentation scheme on DARPA Hub4 Broadcast News (30.5% relative improvement in FES) and NGSW data. Transcript generation is demonstrated for a six-decade portion of the NGSW corpus. Novel model adaptation using structure maximum likelihood eigenspace mapping shows a relative 21.7% improvement. Issues regarding copyright assessment and metadata construction are also addressed for the purposes of a sustainable audio collection of this magnitude. Advanced parameter-embedded watermarking is proposed with evaluations showing robustness to correlated noise attacks. Our experimental online system entitled "SpeechFind" is presented, which allows for audio retrieval from a portion of the NGSW corpus. Finally, a number of research challenges such as language modeling and lexicon for changing time periods, speaker trait and identification tracking, as well as new directions, are discussed in Manuscript

show abstract

Convergence analysis of the quasi‐OBE algorithm and related performance issues

Deller

Gollamudi²,

Nagaraj

et al. 2007

Adaptive Control & Signal

View full text Add to dashboard Cite

Quasi-OBE' (QOBE) is an adaptive set identification and filtering algorithm which is based on the principles of optimal bounding ellipsoid processing, but which has other geometric and classic leastsquares interpretations which greatly enhance its application potential. In particular, because of its unusual optimization criterion, the ellipsoidal membership set associated with QOBE is more likely to retain (i.e. to move in the parameter space with) the system model's 'true parameters,' say h * , when those parameters are time varying. Moreover, in the unlikely event that h * moves outside the set, the integrity of the point-set estimation remains intact, and the estimator provably converges under known conditions. The consistency of the set estimation can be restored at any time using typical 'rescue procedures' if desired. Understanding convergence performance is very critical to successful QOBE application. Convergence analysis of both the central point estimate and measures of the hyperellipsoidal membership set is presented. The main results give conditions for point estimate convergence, and show that set convergence to a point is not possible. Implications of these convergence results for practical application are discussed.

show abstract

Speech watermarking with objective fidelity and robustness criteria

Abstract: ABSTRACT

Cited by 8 publications

References 7 publications

Security of Lattice-Based Data Hiding Against the Watermarked-Only Attack

Security of Lattice-Based Data Hiding Against the Watermarked-Only Attack

SpeechFind: advances in spoken document retrieval for a National Gallery of the Spoken Word

Convergence analysis of the quasi‐OBE algorithm and related performance issues

Contact Info

Product

Resources

About