One of the many things that changed after September 11, 2001 was the speed at which effective translation hardware and software has been developed. In the 1990s there was a lot of progress in computerized language translation and software that could rapidly decipher what someone was saying (so the software and hardware could create the proper translation into another language). Over the last few years this technology has finally matured to the point where there is a lightweight (310 gr/11 ounces) bit of hardware (a smart phone running Android), a miniature speaker, and a headset for the user. It takes about 30 minutes to learn how to use this VRT (Voice Response Translator), which is now called Miltrans for the military version. This version gives the user lists of phrases on the smart phone screen. All the user has to do is press the desired phrase and the foreign language word or phrase comes out on the speaker. The more expensive VRT uses spoken voice input which is subject to some error (about 5 percent) in translating the English word.
VRT provides up to a thousand translatable words or phrases per language and the spoken translation that comes from the speaker comes out clearly in the foreign language. The accuracy of the spoken speech software is 95 percent and it gets better with use. There are phrase sets of about 350 words and phrases for fifty languages so far, as well as software that lets users add phrases. The capacity of the hardware is 1,000 phrases each for up to 199 languages. There is also a megaphone that can be used in place of the small flat wearable speaker. This allows users to be heard in a loud situation. Soldiers have successfully used VRT and Miltrans in raids and situations where there were a lot of agitated and noisy civilians. Foreign aid workers have also found VRT useful in similar situations in disaster zones.
There are over 12,000 VRT and Miltrans units in use and many more may be purchased by military, police, foreign aid, and commercial organizations. VRT and Miltrans are merely the latest in a decade old line of translation devices developed with the assistance of DARPA (Defense Advanced Research Projects Agency), several Department of Defense contracts, and lots of troops willing to try new stuff under combat conditions.
There were several earlier DARPA sponsored translation devices that led up to VRT and Miltrans. Back in 2008 the hot new item was the Phraselator. This was a (large) PDA (or large smart phone) size device that emitted perfectly pronounced words and phrases when the user picked them from a list on the PDA screen or spoke the English version into a microphone. More interestingly, Phraselator had already been around for a decade and had been a lifesaver in Afghanistan and Iraq. The first models stored data on 64 megabyte Compact Flash memory cards. With those early versions you plugged in the right memory card and you are ready to communicate in one of over 50 languages. Each memory card contained 500-1,000 phrases.
Since then, memory card capacity grew to hold a thousand times as much data, enabling all the languages to be stored in each $2,500 Phraselator. Weighing about 450 gr (a pound), the Phraselator was heavy but the battery lasted over a week without a recharge. The voice recognition was pretty good if you spoke clearly and did not require any preparation by the user (like reciting a bunch of phrases so the computer can "recognize" your voice). The phrases in the Phraselator mostly asked questions which could be answered by yes/no, pointing somewhere, or doing something specific (as in, "put your hands up"). Civil Affairs troops liked it because they dealt with people in a relaxed atmosphere where an American with yet another gadget would not produce more anxiety. Medical personnel also found Phraselator very useful.
There were special vocabulary sets in the Phraselator, like Force Protection (FP) and Medic, which made available appropriate terms in many languages like Korean, Dari (Eastern Farsi), Pashtu, Arabic (Gulf), and Urdu. There was a Maritime Intercept Operations (MIO) mode, which was developed for use in boarding operations in the Arabian Gulf. There are over 400 phrases in Arabic, Farsi, Urdu, and Hindi. There was a Debriefing Aid mode that has some 5,000 intelligence debriefing phrases and words rendered in Persian-Farsi, Singhalese, Haitian-Creole, Russian, Serbo-Croatian, Cambodian, Spanish, French, Korean, and Arabic (Egyptian). There were several different medical modes. The one watt speaker on the Phraselator could be easily heard and the unit could be plugged into a public address system as well.
The first Phraselator appeared in 1998, but the true "Phraselator" didn't appear until early 2002, when they were first used in Afghanistan. The Department of Defense had bought nearly ten thousand by 2008, and the troops agreed that it sure beat sign language or thumbing through a phrase book. The locals liked it because it was yet another neat American gadget and one that won't kill them as well.
The next generation of Phraselator was to enable true two-way communication. This technology was already available in 2008 but needed a laptop computer with a big hard drive and a fast CPU. That kind of hardware is now available in smart phones but the two-way Phraselator is still in development. Meanwhile, the hand-held Phraselator is still available but is being replaced by smart phone based units like VRT and Miltrans.