Publications

Vocal fold vibratory asymmetry is often associated with inefficient sound production through its impact on source spectral tilt. This association is investigated in both a computational voice production model and a group of 47 human subjects. The model provides indirect control over the degree of left-right phase asymmetry within a...

Missing or degraded information continues to be a significant practical challenge facing automatic face representation and recognition. Generally, existing approaches seek either to generatively invert the degradation process or find discriminative representations that are immune to it. Ideally, the solution to this problem exists between these two perspectives. To this...

As the scope and scale of Internet traffic continue to increase the task of maintaining cyber situational awareness about this traffic becomes ever more difficult. There is strong need for real-time on-line algorithms that characterize high-speed / high-volume data to support relevant situational awareness. Recently, much work has been done...

This paper presents a usable graphical interface for specifying and automatically enacting access control rules for applications that involve dissemination of data among mobile tactical devices. A specific motivating example is unmanned aerial vehicles (UAVs), where the mission planner or operator needs to control the conditions under which specific receivers...

A traditional trade-off when designing a mission critical network is whether to deploy a small, dedicated network of highly reliable links (e.g. dedicated fiber) or a largescale, distributed network of less reliable links (e.g. a leased line over the Internet). In making this decision, metrics are needed that can express...

The DoD vision of real-time information sharing and net-centric services available to warfighters at the tactical edge is challenged by low-bandwidth and high-latency tactical network links. Secured tactical applications require transmission of digital certificates that contribute a major portion of data in most secure sessions, which further increases response time...

In this paper, we discuss the use of minimum classification error (MCE) training as a means for improving traditional approaches to topic identification such as naive Bayes classifiers and support vector machines. A key element of our new MCE training techniques is their ability to efficiently apply jackknifing or leave-one-out...

A system-on-chip (SOC) implementation is an attractive solution for size, weight and power (SWaP) restricted applications, such as mobile devices and UAVs. This is partly because the individual parts of the system can be designed for a specific application rather than for a broad range of them, like commercial parts...

We present a new perspective on the subspace compensation techniques that currently dominate the field of speaker recognition using Gaussian Mixture Models (GMMs). Rather than the traditional factor analysis approach, we use Gaussian modeling in the sufficient statistic supervector space combined with Probabilistic Principal Component Analysis (PPCA) within-class and shared...

Of increasing importance in the civilian and military population is the recognition of Major Depressive Disorder at its earliest stages and intervention before the onset of severe symptoms. Toward the goal of more effective monitoring of depression severity, we investigate automatic classifiers of depression state, that have the important property...