Award Data

This proposal explores feature representations for automatic speech processing algorithms with a focus on preserving the clients' privacy. We address two different use cases of privacy: a system that transcribes speech, but is unable to infer the identity of speakers, such as an anonymous tip hotline; and a system that identifies speakers, but is unable to recognize the communicated messages, us ...

This award is for participation in the I-CORPS Program through DHS SBIR CRPP funds. Basic requirements include: Attend the initial seminar at the prescribed location (three to four days) Attend the closing lessons learned seminar seven weeks later at the identical location as the opening seminar Carry out one hundred face to face interviews with potential customers.

Fires in tunnels, whether resulting from collisions, bombs, or other sources, are extremely hazardous. Authorities need immediate notification of the problem to provide quick assistance to passengers and prevent additional collisions. One solution is to place temperature sensors along the length of the tunnel; these sensors can be used to detect blasts and issue an alarm. Placing sensors in a tunn ...

Michigan Aerospace Corporation proposes to develop an inexpensive system to sense nitrous oxide (N2O) emissions from agricultural fields using laser-based sensors mounted on drones. These sensors include an optical absorption cell, a wind sensor, and a camera for plant health and ground assessment. The measurements from these sensors will be combined and processed with artificial-intelligence-enab ...

The integration of new cyber technologies in emergency response systems can allow profound improvements in system performance, but also expose the systems to systemic security and privacy risks.Thus, there is a critical need for design-time risk-assessment tools for emergency response systems, which allow evaluation and mitigation of threat impacts across their heterogeneous cyber and physical com ...

In the pattern of life problem space, data is often represented via mathematical graphs, in which a variety of algorithms may be employed to conduct semi-autonomous analysis. While successful empirical application of graph-domain algorithms on ABI problems has been achieved, most of these algorithms require a tuning parameter, which is often set heuristically in real-world scenarios. Arete has dev ...

QUASAR, EyeTracking Inc. (ETI), and Safe Passage International (SPI) will define a closed-loop training system that incorporates EEG and eye tracking technologies to improve Transportation Safety Officer (TSO) training efficiency and effectiveness. To minimize the implementation risk of the Phase II system demonstration, in Phase I the team will: 1. Measure the ability of EEG- and eye-tracking/pup ...

The objectives of the proposed work are to develop the technical feasibility and detailed requirements for new methods and technology to decrease, substantially, the cost and time required to digitize and store high-speed, high-resolution film without sacrificing the spatial and grayscale depth of the original film. The work will be aimed at improvements that would make it practical to digitize a ...

Mass spectrometers necessitate a partial pressure zone to allow for the process of mass characterization of the analyte under consideration. The reasons for transforming neutral molecules into charged ions include (1) the ability to control the trajectories and destinations of charged species by appropriate combinations of electric and magnetic fields and (2) detectability that is generally much g ...

To detect and identify vapors of contraband currency, not to mention a selection mode for explosives, narcotics, chemical warfare agents, biological threats or other analyte species at trace levels in air, is an important problem with many ramifications. One widely used approach is based on transforming the molecules of those species into gas phase ions, which are then characterized by measurement ...