Army's Talking Technology

September 2011
By George I. Seffers, SIGNAL Magazine
E-mail About the Author


Sgt. 1st Class Kristopher Alvey, ARNG, civil military operations noncommissioned officer of the 1st Battalion, 178th Field Artillery, South Carolina Army National Guard, plays with local children near the Walayatee school in Afghanistan. Language translation technologies are expected to help relationships between U.S. forces and local citizens in Afghanistan and in other areas of operation.

The U.S. Army is advancing language translation software.

Military officials in Afghanistan cite the language barrier as one of the most vexing communications obstacles in the battlefield environment. It is a challenge, for example, for U.S. warfighters to communicate effectively with their coalition partners or with the Afghanistan National Security Forces, especially if they are talking over tactical radios during combat chaos. More critical still is the need for warfighters to communicate effectively with Afghan citizens and leaders at all levels.

The U.S. Defense Department recognizes the vital role language translation plays in counterinsurgency operations. Robert Gates, former secretary of defense, issued a memo in May 2010 naming language as one of the seven pillars of the counterinsurgency strategy in Afghanistan, according to Michael Beaulieu, U.S. Army product director for the Machine Foreign Language Translation System (MFLTS). To meet the need, the Army hopes to begin initial fielding of a do-it-all language translation software by January 2015.

MFLTS is expected to be a two-way communications software that can be used on a variety of platforms, including smartphones, laptops and tactical radios. It would work with any operating system, whether it is Windows, Linux or Android. It also is being designed as a free-speech system, meaning it would translate languages on the fly rather than relying on a list of pre-programmed words or phrases. And, it would  translate both spoken language and text. Initial languages likely will include Dari and Pashto, the two languages most common in Afghanistan, but it also would be a modular, plug-and-play system so that additional language upgrades can be added as needed.

Human translators can help forces, but there simply are not enough who can follow U.S. troops through tough terrain or into dangerous situations. And other systems, according to Army officials, offer limited capabilities. Earlier devices sent to the battlefield, for example, used pre-scripted words or phrases, such as “Get out of the car,” or “Do you have a bomb?” A soldier would touch that phrase, and it would come out in the language of the native speaker.

MFLTS aims to solve all translation problems in one fell swoop. “The key thing here is that sometimes people in industry, academia or other government agencies will build gee-whiz, whiz-bang devices that look really cool, but they do only one thing. This is the only program of record within the Defense Department that’s doing machine foreign language translation,” Beaulieu declares. As a program of record, it is a formal Army acquisition effort with defined baselines and milestones that is funded as a program by Congress.

Beaulieu adds that purchasing myriad devices that each solve only one small piece of the puzzle would result in soldiers “carrying around dozens of devices” and would create a “sustainment and maintenance nightmare.”

He reports that MFLTS is destined for eventual deployment on three intelligence systems: the Distributed Common Ground System-Army (DCGS-A), the Prophet Enhanced system, and the Counterintelligence and Human Intelligence Automated Reporting and Collection System (CHARCS). “MFLTS is operating system agnostic, so that means we will be able to plug into other programs of record, regardless of what program they’re using. The intent is that this isn’t a stand-alone system; it’s software that will reside on other program-of-record systems,” Beaulieu says.

The DCGS-A allows analysts to gather data from multiple sensors and distribute it to the Army’s network of battle command systems. Prophet Enhanced vehicles, which have been deployed to Afghanistan, are housed on a Panther variant of the Mine Resistant Ambush Protected vehicle—the armored, six-wheeled vehicle with a V-shaped hull for deflecting blast fragments from improvised explosive devices. Prophet Enhanced carries new, state-of-the-art emitters and upgraded signals intelligence software. It also has fewer computer screens, as well as a graphical user interface, which enables soldiers to find all the information needed on one system. Additionally, the system includes onboard satellite communications, improving the crew’s ability to communicate and conduct missions while away from the forward operating base.

CHARCS, meanwhile, is described in budget documents as providing the Army with automation support for collecting and reporting counterintelligence and human intelligence data to satisfy tactical human intelligence requirements. CHARCS functionality provides support for information collection, reporting, investigation, interrogation, biometrics and document exploitation operations. The CHARCS architecture extends from the individual soldier or counterintelligence agent to theater and national intelligence organizations. CHARCS provides systems to all Army commands, special forces, reserves, National Guard, Stryker Brigade Combat Teams and the training base. CHARCS systems produce and disseminate messages and reports through an array of communications systems, including: Combat Net Radio, Single Channel Ground and Airborne Radio System, Portable Radio Communications (PRC)-150 Secure Telephone Equipment, Secure Telephone Unit, satellite and other organic communications devices. The system also reports collected intelligence directly to operational management teams of U.S. Army intelligence units. Future development efforts will provide counter-intelligence agents and human intelligence collectors with improved collection, reporting, biometrics, language, communications and mission management capabilities.

Beaulieu envisions a wide range of uses. Soldiers could use it, for example, on handheld systems at checkpoints in Iraq and Afghanistan. It also could be used to develop a cultural interface and gather intelligence from local leaders or to enhance coalition operations. “It will benefit greatly—especially in coalition operations—if you’re exchanging operations plans and speaking different languages, or you’re chatting about what’s going on during an operation, or monitoring the local media for open intelligence. Whether you’re building a water project or providing medical care to a local civilian, it’s absolutely critical that you speak the language of the country that you’re in, or communicate with your coalition partners,” Beaulieu insists.

The Army already has fielded some quick-reaction capability technologies to the battlefield to meet an immediate need. Those systems are more limited than the envisioned MFLTS, but the service is gathering feedback and lessons learned that will benefit MFLTS development. One such system, the Foreign Media Monitoring System, offers a real-time automatic machine translation, search and alerting capability across multiple media and languages, providing rapid insight into emerging events. It also allows for the correlation of open-source information and development of intelligence products for strategic or tactical use. Another product, the two-way Speech-to-Speech (2WS2S) translator laptop system, provides a speech translation capability, and the Voice Response Translator provides a hands-free, eyes-free and voice-activated, one-way, phrase-based capability. Beaulieu’s office also fielded the Phraselator P2, which provides voice- and touch-activated one-way phrase-based translation.

Most recently, Beaulieu’s team worked with the Defense Advanced Research Projects Agency and the Army Training and Doctrine Command’s capability manager for biometrics and forensics to provide the 101st Airborne Division with more than 100 2WS2S devices housed on an Android-based smartphone. The devices were sent earlier this year to two battalions within the 4th Brigade Combat Team. The system translates both Dari and Pashto. The operational assessment is expected to be completed this month.

Technology development for MFLTS will take about 15 months. Following that will be the Army’s award of a one-year engineering, manufacturing and development contract followed by a production contract. Fielding could occur in about three years, Beaulieu reports. As of this spring, the Army had $4 million set aside for the program. Initial fielding most likely will be with a Brigade Combat Team and with the CHARCS program office.

Beaulieu emphasizes that the importance of language translation to military operations cannot be undersold. “It may not look pretty on the nightly news, but it helps avoid conflict in-country. For a small amount of money, this technology is an absolute game-changer. People can talk to each other, and you can work out a lot of problems, both with the coalition partners and with the people you’re trying to serve. You can win an awful lot of good will with people when you can talk to them,” he says. “We have a hard enough time communicating with each other in English. To communicate better with our coalition partners or with the people in the country we’re working in is just priceless.”



Enjoyed this article? SUBSCRIBE NOW to keep the content flowing.