Colaboração 2.0: Breaking Down the Language Barrier
In The Hitchhikers Guide to the Galaxy, Douglas Adams famously dreamed up the babel fish, a “small, yellow, leech-like” creature, “probably the oddest thing in the universe.” Feeding on brainwave frequencies, the babelfish telepathically excretes understandable speech; when implanted in the brain through the ear, the babel fish allows its host to instantly understand any form of language.
While the discovery of a benevolent brain parasite is unlikely, universal translation is quickly becoming more fact than fiction. Completely fluent translations are still around the proverbial corner, but innovations in computer translations over the last decade have brought us much closer to computer-aided polyglotism. Today’s technology already suggests significant implications for business and personal communication.
The first commercial translation software programs became available in the 1990s, including one from Alta Vista, in fact called Babelfish. (It was later sold to Yahoo!, and now redirects to Bing Translator.) However, effective translation requires computers to make difficult leaps of computer logic, and it wasn’t until recently that these tools have been able to do that with any sophistication.
Google Translate emerged in 2007 as the first free statistics-based, “machine translation” service. Its translation program draws on a massive language database from which it can identify the most probable sentence constructions and analyze how they coincide across languages. The site now provides instant translation of inputted text, web pages, and uploaded documents in 64 different languages. Google’s model is outperforming traditional grammar-based translation programs, and has become a tremendously popular tool. As Google research scientist Frank Och recently blogged: “What all the professional human translators in the world produce in a year, our system translates in roughly a single day. By this estimate, most of the translation on the planet is now done by Google Translate.”
The program can be very useful when composing basic e-mails to foreign language speaking team members or potential business partners abroad. But more complicated sentence structures and context-based descriptions are beyond its capacity, as many high school French students with failing homework grades assuredly now know. For instance, if these students wanted to say “I can hardly speak any french at all!”, in French, they might use the common French expression “Je parle le francais comme une vache espangole!” But inputted in Google translate, you’d get the literal translation “I speak french like a spanish cow,” which would be confusing, minus the benefit of cultural context.
Professional translation services, then, are still needed to handle formal documents, especially contracts and other business agreements. But Google Translate and other sophisticated language services could well open the door to more frequent exchanges across the once-vast language divide.
This is especially true now that Google Translate can be integrated into both Firefox and (Google’s own) Chrome browser, allowing for instantaneous webpage translation. Investors looking to capitalize on Brazil’s market size, growing middle class, and hosting honors for the upcoming World Cup and Summer Olympics can peruse the business section of the country’s top newspaper, O Globo. Telecommunications firms interested in Gabon’s numerous infrastructure development programs can switch the government’s official website from French to English or Chinese.
As impressive as this technology might seem, innovation in computerized translation is not limited to text. AT&T recently announced that it would release its Watson API to the open source community this month. Watson’s capabilities are impressive: speech in one language can be converted to text in near real time, followed by a text translation (with little delay) and playback by a computer generated voice .
The free Google Translate mobile app also provides speech-to-speech translation, and the the Android version also accepts handwritten input. Not to be outdone, Microsoft has developed a telephone translation app for Windows phone that works through the phone’s camera to translate printed text and through the phone’s voice recognition software to translate spoken text.
A number of impressive translation pilot projects debuted at the most recent TAUS User Conference, including a tablet application that can help health care workers communicate with foreign language-speaking patients, and real-time chat translation services that can help call centers (both domestic and international) communicate more efficiently with foreign customers.
The rapid advances of the past few years suggest even more impressive breakthroughs in the near future. While translation software will likely always have shortcomings—especially with regard to cultural issues, idioms, and slang—the effective translation of even half of written and spoken communications situations could completely revolutionize the way the global market does business.
Do you have any favorite tools to break down language barriers? Tips for collaborating with foreign-language speaking project team members? Favorite quotes from Hitchhiker’s Guide? Let us know below in the comment section!