Hacker News Reader

Comment by sofon | Hacker News Reader

I'm curious as to why automated translation seems to be so bad for Japanese<->English.

Is Japanese really such an outlier?

English <-> Chinese seems more reasonable. Translation between European languages I would guess is an almost solved problem, but how about between language groups in general?

View on HN

It could be cultural, in that Japanese communication style can often be understated, with indirect implications pointing at a meaning without directly specifying it. That can vary greatly depending on the context and subject matter, though.

Ouch you tell me!! A couple of our customers use high tech products manufactured in China and Japan, for which the manuals were badly translated from the native language to English, and then we have to mop it up from English to French, for example. It is very labor intensive to decipher what the intent is and then to translate it into direct French sentences, in the style most readers are expecting. Thanks for pointing it out! Again, I doubt a machine could do that kind of work.

As a general rule it probably depends on what you're trying to translate. If it matches the corpus well, you'll do better. This is simplified, of course, as NMT is a bit of a black-box. But we know the GIGO concept still applies.

Perhaps Japanese has a smaller corpus to draw from. Perhaps less work has been done in that pair. Perhaps you're just not translating the right kind of text. It's hard to say.

Comparing Japanese to Chinese is not really apples to apples. Likewise for European languages. Sharing a writing system in any form (Latin/Roman or Chinese) doesn't really come into it.

Reply on news.ycombinator.com