The Future of Ubiquitous Spoken Content with Douglas W. Oard

The final part of the Good Systems: Acoustic Surveillance & Big Data Series. Today, well over half the world’s conversations occur in the presence of a device capable of capturing and recording what is spoken.  Storage capacity is being built far faster than would be needed to record and retain any of that content, network capacity to move recorded content to central storage is widely available, and speech recognition techniques are now sufficiently accurate to render searchable much of the spoken content that is recorded.  Depending on your perspective, this could provide either an unprecedented bounty of information that could be mined to create new services with the potential for great benefit, or a looming threat to privacy that we must control before it can be used to control us.  Of course it is both.  In this talk I will review the technologies that have brought us these challenges and opportunities, and I’ll identify some remaining technical challenges that currently limit their reach and application.  I will then focus the remainder of my remarks on the interplay between technology and policy that will shape the ways in which we might seek to achieve a balance between the risks and benefits that this cornucopia of new information could offer.

Douglas Oard is a Professor and Associate Dean for Faculty at the College of Information Studies at the University of Maryland, College Park.  An electrical engineer, his research interests include the design and evaluation of technology for searching human language; examples include speech retrieval, cross-language retrieval, and document image retrieval.  Additional information is available at

Friday, April 12, 2019 at 9:00am to 12:00pm

Peter O'Donnell Jr. Building (POB), Vislab
201 24TH ST E, Austin, Texas 78712

Event Type

Science & Tech


Office of the Vice President for Research, Good Systems

Target Audience

Students, Staff, Faculty

Google Calendar iCal Outlook

Recent Activity