DARPA awards Raytheon BBN Technologies to advance automatic translation

Raytheon Company's BBN Technologies, based in Cambridge, MA, has been awarded an additional $6.14 million in funding by the Defense Advanced Research Projects Agency (DARPA) under the Multilingual Automatic Document Classification, Analysis and Translation (MADCAT) program.

The MADCAT program's goal is to create a prototype system that quickly provides accurate, relevant, distilled, actionable information to military commands and personnel. It does this by converting foreign language text images, such as road signs, flyers, photographs, and handwritten notes, into English transcripts automatically, without the use of linguists and analysts.

Prem Natarajan, head of speech and language processing at Raytheon BBN Technologies, said, "Foreign language translation on the battlefield is slow, dangerous and expensive. The MADCAT system will help our troops understand road signs, print media and captured documents that could be of immediate importance to their safety and to the successful completion of their missions."

Under the contract, Raytheon BBN Technologies will further refine a prototype translation system developed under previous awards in this program that can be deployed on a laptop computer. To accomplish this, Raytheon BBN Technologies is integrating optical character recognition with its state-of-the-art translation and distillation techniques and developing novel methods to process handwritten text.

This award follows Raytheon BBN's success in meeting required milestones during the first two years of the MADCAT program.