June 27, 2018
Attendance is limited to JHU/APL employees with a SECRET clearance.
In this classified talk, David Taubenheim describes some of the challenges facing the National Intelligence Community Program Area’s sponsors in processing a large volume of audio data collected from a variety of imperfect sources. With the goal of accelerating the intelligence analysts’ and linguists’ workflow, the suitability and limitations of commercial human language technologies (HLT) are examined, as is the application of customized HLT tools developed both internally and externally to APL. Advances in machine learning for rapid audio triage and tagging are examined from the perspective of a sponsor’s mission. The focus of the talk then shifts to technologies that may set a new bar of subterfuge in audio collection and human speech in the very near future. Audio clips for the collection and processing technologies are featured.
David Taubenheim is the Program Area Manager of JHU/APL’s National Intelligence Community program area, a team of engineers, scientists, and technical managers who support US Government sponsors through specialized technology projects and emergent initiatives. Before serving in the program office, David joined JHU/APL’s custom integrated circuit team in 2009 to engineer embedded digital signal processing solutions for miniaturized applications. Between 1996 and 2009, David worked at Motorola, where the majority of his effort involved the design of communications systems using customized software-defined radio platforms, earning him the company’s Distinguished Innovator Award in 2008. While there, David also moonlighted as an instructor for Xilinx, Inc., teaching digital signal processing design techniques using field programmable gate arrays. At Motorola, he contributed to the then-nascent ZigBee and IEEE 802.15.4 low-power wireless IoT network standards by working as part of a design team to create the first 802.15.4 compliant radio transceiver chip in 2002.
David graduated from the University of Illinois at Urbana-Champaign with a Bachelor’s of Science in Electrical and Computer Engineering in 1996 and completed his Masters of Science in Electrical Engineering from the National Technology University in 2001, studying during his employment at Motorola. He has been granted 16 patents, primarily in the area of wireless communication. In 2018, he was presented with the LGBTQ+ Engineer of the Year Award by the National Organization of Gay and Lesbian Scientific and Technical Professionals.