The award was given by the U.S. Defense Advanced Research Projects Agency under the Multilingual Automatic Document Classification, Analysis and Translation program, under which the company has already been working.
"Foreign language translation on the battlefield is slow, unreliable, difficult to access and expensive," said Prem Natarajan, head of Speech and Language Processing at Raytheon BBN Technologies.
"By putting translation capability directly in the hands of the end user, MADCAT technologies will help our troops understand handwritten and printed documents that could be of immediate importance to their safety and to the successful completion of their missions."
Work under the award will include advancing previous company efforts to refine a laptop-deployable prototype translation system that will automatically convert foreign language text images -- handwritten notes and machine-printed documents, for example -- into English transcripts without the use of linguists and analysts and integrate optical character recognition with translation techniques.