Aparna Gurijala

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

In this study, we discuss a number of issues for audio stream phrase recognition for information retrieval for a new National Gallery of the Spoken Word (NGSW). NGSW is the first largescale repository of its kind, consisting of speeches, news broadcasts, and recordings that are of historical content from the 20 th Century. We propose a system diagram and discuss critical tasks associated with effective audio information retrieval that include: advanced audio segmentation, speech recognition model adaptation for acoustic background noise and speaker variability, and natural language processing for text query requests. A number of questions regarding copyright assessment, metadata construction, digital watermarking must also be addressed for a sustainable audio collection of this magnitude. Our experimental online system entitled “SpeechFind” is presented which allows for audio retrieval from a portion of the NGSW corpus. We discuss a number of research challenges to address the overall task of robust phrase searching in unrestricted audio corpora. 1. Overview The problem of reliable speech recognition for spoken

Speech watermarking with objective fidelity and robustness criteria

Speech watermarking strategies inevitably alter original signal content. Fidelity is adversely affected by increased perturbation while watermark robustness to attack is generally improved. Parameter-embedded watermarking is effected through slight perturbations of parametric models of some deeply integrated dynamics of the speech. Within this framework, a specific algorithm is presented in which the fidelity-robustness tradeoff can be objectively assessed and quantifiably adjusted according to specific measures. An overview of the general parameter-embedding strategy is followed by presentation of the featured algorithm, analysis of its properties, and experiments with speech data to assess fidelity, robustness, and other performance properties.

On encryption-compression tradeoff of pre/post-filtered images

Advances in network communications have necessitated secure local-storage and transmission of multimedia content. In particular, military networks need to securely store sensitive imagery which at a later stage may be transmitted over bandwidth-constrained wireless networks. This work investigates compression efficiency of JPEG and JPEG 2000 standards for encrypted images. An encryption technique proposed by Kuo et al. in [4] is employed. The technique scrambles the phase spectrum of an image by addition of the phase of an all-pass pre-filter. The post-filter inverts the encryption process, provided the correct pseudo-random filter coefficients are available at the receiver. Additional benefits of pre/post-filter encryption include the prevention of blocking effects and better robustness to channel noise [4]. Since both JPEG and JPEG 2000 exploit spatial and perceptual redundancies for compression, pre/post-filtered (encrypted) images are susceptible to compression inefficiencies. The PSNR difference between the unencrypted and pre/post-filtered images after decompression is determined for various compression rates. Compression efficiency decreases with an increase in compression rate. For JPEG and JPEG 2000 compression rates between 0.5 to 2.5 bpp, the difference in PSNR is negligible. Partial encryption is proposed wherein a subset of image phase coefficients are scrambled. Due to the phase sensitivity of images, even partial scrambling of the phase information results in unintelligible data. The effect of compression on partially encrypted images is observed for various bit-rates. When 25% of image phase coefficients are scrambled, the JPEG and JPEG 2000 compression performance of encrypted images is almost similar to that of unencrypted images for compression rates in the 0.5 to 3.5 bpp range.

Why watermark?: the copyright need for an engineering solution

An important research component in the creation of the National Gallery of the Spoken Word (NGSW) is the development of watermarking technologies for the audio library. In this paper we argue that audio watermarking is a particularly desirable means of intellectual property protection. There is evidence that the courts consider watermarks to be a legitimate form of copyright protection. Watermarking facilitates redress, and represents a form of copyright protection that universities can use without being inconsistent in their mission to disseminate knowledge.

Robust algorithm for watermark recovery from cropped speech

Most digital watermarking techniques are susceptible to damage by data cropping. Although the effects of cropping might not be perceptible, watermark recovery may be rendered difficult or impossible due to the desynchronization of the recovery process. The transform encryption coding (TEC) based watermarking algorithm was presented at ICASSP 2000 (see Ruiz, F.J. and Deller, J.R., Jr, IEEE International Conference on Acoustics, Speech and Signal Processing 2000). This paper investigates the performance of TEC watermarking in the presence of cropping, and presents an algorithm that identifies cropped samples and recovers watermarks from the damaged record. Implementation details and experimental results under different environmental conditions are presented.

Audio watermarking techniques for the National Gallery of the Spoken Word

This is one of two companion papers describing technical challenges faced in the development of the National Gallery of the Spoken Word (NGSW). The present paper describes watermarking technologies for intellectual property protection. Following an introduction to data watermarking, the paper focuses on a new algorithm called \textit{transform encryption coding} (TEC) and its application to watermarking the NGSW archives. TEC has a number of flexible features that make it amenable to the NGSW development.

On the robustness of parametric watermarking of speech

Parameter-embedded watermarking is effected through slight perturbations of parametric models of some deeply-integrated dynamics of a signal. This paper is concerned with particular model form, linear prediction (LP), which is naturally suited to the application of interest, speech watermarking. The focus of this paper is on the robustness performance of LP-embedded speech watermarking. It is shown that the technique is quite robust to a wide array of attacks including noise addition, cropping, compression, filtering, and others. In the LP formulation, a set-theoretic adjunct to the parameter embedding can be used to identify a watermark that is optimally robust against certain attacks, within a quantified fidelity constraint.

Detector Design for Parametric Speech Watermarking

Parametric watermarking is effected by modifying the linear predictor coefficients of speech. In this work, the parameter noise is analyzed when watermarked speech is subjected to additive white and colored noise in the time domain. The paper presents two detection techniques for parametric watermarking. The first approach uses the Neyman-Pearson criterion to solve a binary decision problem. In the second approach, discriminant functions based on the minimum-error-rate criterion are used to determine which one of the many watermarks was embedded or if no watermark is present. Experiments with speech data are used to determine the false-alarm and missed detection rates

Robustness optimization of parametric speech watermarking

Parameter-embedded watermarking is effected through slight perturbations of parametric models of some deeply-integrated dynamics of the speech signal. In set-membership filtering (SMF) based parametric watermarking, linear predictor (LP) coefficients of the original speech are modified subject to an objective fidelity constraint. SMF is used to obtain sets of allowable parameter perturbations (i.e., watermarks) subject to a constraint on the error between the watermarked and original material. This paper discusses the robustness of SMF based watermarking to filtering, quantization and combination attacks. An important consideration in watermark robustness is the energy of the watermark signal (difference between watermarked and original signals). The most robust watermark is obtained from perturbed LP coefficients at the boundary of the membership set. A constrained optimization problem is solved to obtain the best watermarks for filtering and quantization attacks

Assessment of camera phone distortion and implications for watermarking

The paper presents a watermark robustness model based on the mobile phone cameras spatial frequency response and watermark embedding parameters such as density and strength. A new watermark robustness metric based on spatial frequency response is defined. The robustness metric is computed by measuring the area under the spatial frequency response for the range of frequencies covered by the watermark synchronization signal while excluding the interference due to aliasing. By measuring the distortion introduced by a particular camera, the impact on watermark detection can be understood and quantified without having to conduct large-scale experiments. This in turn can provide feedback on adjusting the watermark embedding parameters and finding the right trade-off between watermark visibility and robustness to distortion. In addition, new devices can be quickly qualified for their use in smart image applications. The iPhone 3G, iPhone 3GS, and iPhone 4 camera phones are used as examples in this paper to verify the behavior of the watermark robustness model.