In "Star Trek," the "universal translator" allowed members of alien species to talk directly without a lot of awkward pauses. It saved the writers grief and gave Capt. Kirk a tactical advantage.

The Army isn't there yet, but it recently took a step in that direction. The service rolled out the Machine Foreign Language Translation System, or MFLTS, system to some 700 users. Army leaders plan this summer to deploy the software across the Project Manager Distributed Common Ground System — Army (PM DCGS-A) portfolio.

Given the Army's expanding expeditionary role, "there is no way to train a sufficient number of linguists in so many relatively uncommon and obscure languages," said Mike Doney, the product manager for MFLTS. "So we are trying to augment the current two-legged capability with a modest level of capability to as many soldiers we can."

The system combines voice recognition and a speech synthesizer to enable on-the-fly translation. The DCGS-A version will incorporate the ability to translate scanned text. Presently the software can accommodate the two spoken languages most commonly encountered by Army personnel in the field, Iraqi Arabic and Pashto.

While two users can converse fluently through the system, it was not designed to deliver a high level of sophistication. With conversational fluency defined on a zero-to-five scale, human linguists operate at a four, whereas MFLTS works at a one.

"You are able to convey rudimentary concepts between two speakers. It achieves basic communication," Doney said.

That may very often prove sufficient for soldiers' needs, said Tracy Blocker, MFLTS product director lead within Army Training and  Doctrine Command's Capability Manager-Biometrics and Forensics at Ft. Huachuca, Arizona.

"At a checkpoint you can say: 'Do you have any weapons? Get out of the vehicle.' If the foreign language speaker complies with that, we consider that a success. Our currently released product meets those requirements," Blocker said.

Native support

The translator has been through extensive field testing, including a February war fighter experiment at Ft. Benning.

"Soldiers would interact with a foreign language speaker who had some relevant information and they would ask questions," Blocker said. "We had native Iraqis there supporting our exercise and it worked well. They were able to ascertain information that aided them with their situational awareness and understanding so they could better prosecute their mission."

While that one-to-one capability is significant, planners say the emerging text translation capability will also play a critical role.

"There is a lot of information that needs to get looked at. There are a lot of things intelligence analysts want to use to contribute to situational understanding of the operational environment," Doney said. These might be papers captured in the field, or they may include electronic documents taken from web pages or social media screen shots. "Whatever the source data is, the text to text translator will process that input and put it out in English."

The paper-translation capability has proven an engineering challenge. "The hard task is ingesting the hard copy document through some scanning or photographic process that yields a clear enough image that an optical capability could turn that image into text," Doney said. "We have not yet fully deployed that capability, but we are very close."

The translator relies on machine learning, a sub-discipline within the general field of big data analytics. The development team will be leaning heavily on that capability as it aims to roll out speech and text translation capabilities for more than 60 languages over the next several years, with the first expansions set to come in the next year or two. On deck are Dari, Farsi, Levantine Arabic, Kurdish, Urdu and Russian.

Of the remaining languages, "a lot of them are dialects of Arabic, but they are all so different from one another, they behave like different languages, so soldiers could lose their ability to

communicate as they move from one area to another if we didn't build those out," Doney said.

The end goal is a massive online downloadable database, which would enable soldiers to select from a portfolio of language packages, downloading from a web portal as needed.

Share:
More In C2/Comms