… experts describe the project as the most ambitious undertaking in the history of human language technology.
If it is developed as planned, the first-of-its-kind machine will be able to recognize speech in multiple languages, translate it into English, and then mine the resulting transcripts to sift so-called intelligence from dross, said sources close to the project.
The ultimate goal of the endeavor, dubbed GALE for Global Autonomous Language Exploitation, is to turn the staggeringly large volumes of recorded foreign language broadcasts, phone conversations, and Internet traffic into something national security analysts, spooks, and soldiers can actually use.
It is said that the National Security Agency gathers enough information every hour to fill the Library of Congress. Most of it is never translated, and never reaches the desk of an analyst.
The dearth of solid human language technology in the hands of the government is a â€œhuge problem,â€? has said Gilman Louie, the president and CEO of In-Q-Tel, the Central Intelligence Agencyâ€™s venture capital arm…
DARPAâ€™s ambitions for GALE represent the â€œHoly Grailâ€? of human language technology, said John Makhoul, a scientist at BBN working on the GALE project. The best speech recognition software operating in a controlled environment like a television broadcast can usually get nine out of 10 words correct.
DARPA wants 95 percent accuracy rates for both speech recognition and translation. And it demands that the engines be able to process radio, TV, talk shows, newswires, newsgroups, blogs, and phone conversations in English, Arabic, and Chinese.
The Pentagon is seeking the same high accuracy rate for the translation part of the system. Experts say that translation software now performs with around 80 percent accuracy.
Archive for August, 2005
Schoenberg Center for Electronic Text & Image
University of Pennsylvania Library
The University of Pennsylvania Library is presently seeking a bright, creative individual with a solid background in humanities computing to guide and manage its electronic publishing unit, The Schoenberg Center for Text & Image (SCETI).
SCETI (http://dewey.library.upenn.edu/sceti) was created in 1996 to produce virtual facsimiles of rare and special materials from Penn’s collections. Over the years, SCETI has evolved into a fully integrated electronic library containing digital versions printed books, manuscripts, correspondences, images, and most recently recorded sound.
SCETI’s projects have ranged from Shakespeare, medieval Judaica, and the traditions of alchemy to ancient papyri, the works of Theodore Dreiser, and the spoken word. In 1998 SCETI received an NEH Challenge Grant, which was successfully met. SCETI collaborates on the Penn campus with faculty and academic units and internationally with institutions across North America and beyond. The SCETI web site is a destination stop for humanities scholars from around the world.
Potential candidates are invited to submit a letter of application which addresses the needs and qualifications of the position, along with their resume and the names, addresses and phone numbers of three references who can address the suitability of the candidate for the position described, to: Robert Eash, Human Resources Officer, University of Pennsylvania Library, 3420 Walnut Street, Philadelphia, PA 19104-6206 or email to: firstname.lastname@example.org
(seen on Humanist)
Interesting comments on measures of success for Wikipedia by Jimbo Wales, guest-blogging at Lawrence Lessig’s site:
So, how are we doing? What are the odds of this goal being accomplished in the next 20 years?
First, it can be argued that although much work remains to be done in many areas, if you speak English, German, French, or Japanese, and have broadband Internet access, you have your encyclopedia. Each of those 4 languages has more than 100,000 articles and provides a reasonably comprehensive resource. Several other languages will pass the 100,000 threshold soon enough, and in 5 years time, all of these and many more will be larger than 250,000 articles.
Second, clearly there is a lot of work to be done in finding ways to actually distribute the work we have done already into areas where people can use it. Many people would be able to make positive use of English, French, or Spanish Wikipedia (for example) if only they had access to it.
Third, while it is important to provide our work in important global or “colonial” languages, we also think it is extremely important to provide our work in languages that people speak natively, at home. (Swahili, Hindi, etc.)
I will define a reasonable degree of success as follows, while recognizing that it does leave out a handful of people around the world who only speak rare languages: this problem will be solved when Wikipedia versions with at least 250,000 articles exists in every language which has at least 1,000,000 speakers and significant efforts exist for even very small languages. There are many local languages which are spoken by people who also speak a more common international language — both facts are relevant.
I predict this will be completed in 15 years. With a 250,000-article cutoff, English and German are both past the threshold. Japanese and French will be there in a year. Several other languages will be there in two years.
The encyclopedia will be free.
More on the potential of games in education, from James L. Morrison, Editor-in-Chief, Innovate:
The August/September 2005 of Innovate’s special issue on the role of video game technology in educational settings is now available at
Innovate is a peer-reviewed, bimonthly e-journal published as a public service by the Fischler School of Education and Human Services at Nova Southeastern University. It features creative practices and cutting-edge research on the use of information technology to enhance education.
Jim Gee opens the issue with a key question: “What would a state of the art instructional video game look like?” Gee’s response focuses on the commercial game Full Spectrum Warrior in order to reveal the “good theory of learning” that should inform the design of video games produced specifically for instructional purposes. In turn, David Shaffer elaborates a similar theory of situated and action-based learning with the concept of an “epistemic game,” whose design integrates player interests, domain knowledge, valued professional practices, and assessment to generate motivation and deep learning. In the following article, Richard Halverson reinforces the argument that valid learning principles inform successful video games, and describes how they might be integrated in educational contexts.
Melanie Zibit and David Gibson report the work in progress on simSchool–a video game that prepares teachers for the complexities of classroom management by offering a “simulated apprenticeship” that prepares teachers to practice the kind of informed decision making required for success in their profession.
Kurt Squire’s findings about the benefits of and obstacles to the implementation of video games in the classroom are based on his own attempt to use Civilization III in high school history classes. He argues that, rather than thinking about how to design good games for the existing K-12 educational system, we should focus our energies on how to design an educational system flexible enough to accommodate video games. In contrast, Michael Begg, David Dewhurst, and Hamish Macleod advocate a “game-informed learning” approach that would make conventional learning activities more game-like. The two medical simulations they describe immerse students in a professional identity and generate highly motivated constructivist learning.
In a provocative glimpse into the future learning landscape, Joel Foreman, this issue’s guest editor, interviews Clark Aldrich, described by Fortune magazine as one of the top three e-learning gurus. The interview begins with the distinction between games and simulations and concludes with Aldrich’s “20 simulations” approach to the reformation of education.
Stephen Downes wraps up the issue with his review of Apolyton, an exemplar site that provides both fodder for resourceful students and models for educators who want to cultivate new online learning communities.
We hope that you enjoy this special issue of Innovate. Please use Innovate’s one-button features to comment on articles, share material with colleagues and friends, easily obtain related articles, and participate in Innovate-Live webcasts and discussion forums. Join us in exploring the best uses of technology to improve the ways we think, learn, and live.
Please forward this announcement to appropriate mailing lists and to colleagues who want to use IT tools to advance their work.
Finally, if you wish to continue to get announcements of new issues, please subscribe to Innovate at www.innovateonline.info Subscription is free.
Several articles collected by CIT:
“Meet the Gamers” By Kurt Squire & Constance Steinkuehler Library Journal.com, April 15, 2005
“In the past, librarians have often been perceived as gatekeepers, arbiters of access to information. The digital cultures now emerging (with the help of technologies such as games) suggest that the days for such an institutional role are numbered.”
“Gaming for Librarians: An Introduction” By Heather Wilson VOYA, February 2005
“[P]eople lament the fact that teens are playing video games and not reading. They are missing the point. Gaming often requires reading, problem-solving, and critical thinking.”
“The Games Children Play” 30-minute video produced by School Matters
Video advocating the use of games in education, featuring interviews with Henry Jenkins, director of comparative media studies at the Massachusetts Institute of Technology and Jim Gee, professor of education at the University of Wisconsin-Madison.
“Video Games and the Future of Learning” By David Williamson Shaffer, Kurt R. Squire, Richard Halverson, and James P. Gee December 2004
“[W]e describe an approach to the design of learning environments that builds on the educational properties of games, but deeply grounds them within a theory of learning appropriate for an age marked by the power of new technologies. We argue that to understand the future of learning, we have to look beyond schools to the emerging arena of video games.”
In a public announcement from the NEH:
Austin, University of Texas, Austin (Collaborative Research) $100,000
Project Director: Michael Gagarin
Project Title: The Laws of Ancient Crete
Description: Publication, both in print and online, of the Greek texts of extant laws from ancient Crete, along with translations and commentaries. (12 months)
Charlottesville, University of Virginia (Collaborative Research)
Outright: $80,000, Additional Match: $20,000
Project Director: Bernard Frischer
Project Title: Aquae Urbis Romae: The Waters of the City of Rome
Description: Adding GIS data and expanding a freely accessible and fully interactive inventory of Roman hydraulic infrastructure from the early Christian era through the early modern period. (24 months)
The Eton Greek Software Project offers now an online language tool for anyone who is learning classical Greek. On offer is a variety of programs which allow the user to test their knowledge of grammar and vocabulary (some require a Java-enabled browser or use Flash). This covers material from the course text ‘Reading Greek’ (one of the most popular Greek courses in higher education) as well as from AS level and GCSE syllabi. Tests can be customised to suit the user’s requirements, according to level of difficulty, English to Greek or vice versa, length of time allowed, or types of words required. There is also an accidence tester designed for practising verb conjugation. A copy of the Eton Greek word list (with translations) for AS level is available to download too; this contains key vocabulary which all Greek learners should know. The testers themselves, however, cannot be accessed via the Project’s homepage – the user must first access Eton College’s homepage and then enter the site via ‘Eton in Action’/’Greek Project’.
(Thanks to the Humbul Humanities Hub)
The Department of Theatre, Whitman College, is now hosting the The Virtual Reality Theatre Tour site, offering visual tours of ancient theaters via panoramic views from several points of view. While still a work in progress (not all listed theaters are linked to a tour yet), it already offers a series of theatres, particularly from Asia Minor (Aspendos, Aphrodisias, Bodrum, Ephesus, Hierapolis, Miletus, Pergamon and Priene). The individual tours can be accessed by clicking on the appropriate location on a linked map. Each tour is accompanied by details including: information on the location of the theatre; dates of construction and renovation; dimensions; and brief details of excavations. Plans of the theatres, and in some cases reproductions, are also given, and there is a glossary of relevant architectural terms. QuickTime software is required to view the virtual tours.
(Thanks to Humbul Humanities Hub)