Thursday, May 15, 2008
...and that's great news any way you say it. Language is one of the biggest challenges we have in making information universally accessible. As part of the machine translation team within Google Research, I'm happy to report we've been hard at work to overcome this challenge. We've recently added translation capabilities for 10 new languages to Google Translate, bringing the total to 23 languages. The newly featured languages include Bulgarian, Croatian, Czech, Danish, Finnish, Hindi, Norwegian, Polish, Romanian and Swedish.
In addition, you can now translate text and web pages as well as perform cross-language searches between any two languages that we offer. For example, we now support Chinese translation to/from any of our languages (e.g., Chinese to French). So for those of you who will be following or attending the Olympics in Beijing this fall, you'll be able to more easily find and access content from local sources.
We've also added a "Detect Language" option to help you automatically identify the language of the text you're trying to translate. Keep in mind that the longer the text, the more accurate it will be. And for those of you who have embedded the Google Translate My Page gadget in your website to give it global reach, these new languages will automatically appear. Developers can also take advantage of these new languages in our AJAX language API.
While our system is quite good, we know it's not perfect. Machine translation is a hard problem, but it plays an important role in helping people access content they might otherwise be unable to read. We’re constantly working to improve the quality, so if you find a translation that’s not quite right, let us know by using our "Suggest a better translation” feature.