Machines Can Ease the Olympics Translation Crunch

The Olympics is a boon to translators. Much of the reporting, interpretation and documentation for the massively international event is handled by humans, but human translators with the right skills can be scarce. “Between some pairs of languages, there are very few people who are experts in both,” said Sanford Cohen, founder of message translation firm SpeakLike. But it’s not just the languages needed, either. New forms of communication like IM, email, voicemail and the web demand different approaches, and computers can help with both challenges.

Machine translation is nothing new: Systran, founded in 1968 to help translate Cold War communications, powered the 1997 launch of the Babelfish service that popularized online translation, and until recently, it was behind Google’s translation systems. Humorous results aside, machine translation works well when software has access to sample text or past translations. “There has been a significant improvement in translation quality because of computing power,” Dimitris Sabatakakis, Systran’s CEO, said.

Read More about Machines Can Ease the Olympics Translation Crunch

Google Translation Center: The World’s Largest Translation Memory

Disclosure: I am the founder of Der Mundo, a multilingual blogging service and translation community that combines human and machine translation (provided in part by Google), and I have researched translation technology for more than 10 years via the Worldwide Lexicon project.

Blogoscoped reports that Google is preparing to launch Google Translation Center, a new translation tool for freelance and professional translators. This is an interesting move, and it has broad implications for the translation industry, which up until now has been fragmented and somewhat behind the times, from a technology standpoint

Google has been investing significant resources in a multi-year effort to develop its statistical machine translation technology. Statistical MT works by comparing large numbers of parallel texts that have been translated between languages and from these learns which words and phrases usually map to others — similar to the way humans acquire language. The problem with statistical MT is that it requires a large number of directly translated sentences. These are hard to find, and because of this SMT systems use sources like the proceedings from the European Parliament, United Nations, etc. Which are fine if you’re writing in bureaucrat-speak, but aren’t so great for other texts. Google Translation Center is a straightforward and very clever way to gather a large corpus of parallel texts to train its machine translation systems.

Read More about Google Translation Center: The World’s Largest Translation Memory