Saturday, March 06, 2010

Google translate 2010 - a marker

This is the Feb 2010 Google Translate output from a Chinese language original ...
天涯周立波

... Perpetual motion machine needs rotor, the rotor is Zhao Benshan and Guo Degang. Zhou Libo positioning themselves with the middle-class waiter, while a decorated themselves with the middle class, followed by a run did not forget about Zhao Benshan. Said Zhao Benshan services for farmers, location of China's rural areas. With that, make up an "In fact, I always have great respect for Old Zhao" and then express their hair a period of "cultural pluralism", the Old Zhao has also stepped on, their body position also climbed up. The results we all know that Old Zhao's fans will feel that "feelings were seriously damaged," the rotor to turn up...
Some of it can be understood, other bits read like receptive aphasia word salad. Amusing, but not terribly informative.

I'm keeping this example as a baseline for the translation results of 2011.

Update 3/21/2010: A NYT OpEd by a translator makes light of Google's efforts.
... Google Translate is a statistical machine translation system, which means that it doesn’t try to unpick or understand anything. Instead of taking a sentence to pieces and then rebuilding it in the “target” tongue as the older machine translators do, Google Translate looks for similar sentences in already translated texts somewhere out there on the Web. Having found the most likely existing match through an incredibly clever and speedy statistical reckoning device, Google Translate coughs it up, raw or, if necessary, lightly cooked....
The approach seems to work well for English/French, but it fails miserably for English/Chinese. The sentences translated from Chinese seem individually meaningful, but the paragraphs are nonsensical.

The critic sounds frightened to me -- and he should be. Google translator works quite well for similar languages. On the other hand, they have another great leap to take if they're going to bridge the Chinese to English gulf.

No comments: