Statistical Machine Translation Gets Real: A Profile of Language Weaver

Statistical machine translation is an innovative way of automatically translating text from one language into another. It’s being used by Google, it’s being used in Iraq, and it’s being commercialized by a company called Language Weaver.

I profiled this company for the December issue of Global by Design and now we’re making the full article available for free download.

We’re going to be hearing a lot more about this technology and Language Weaver because companies can only afford to translate a fraction of their content using translators. SMT is not designed to put translators out of work but instead unleash vast amounts of content that would never have been translated in the first place.

If you want to get an idea of what SMT can do, what it can’t do, and why I think it’s going to revolutionize the translation industry, check out the article.

download article (2.1MB)

(Visited 2 times, 1 visits today)

Author: John Yunker

John co-founded Byte Level Research in 2000 and is author of The Web Globalization Report Card. He also co-founder of Ashland Creek Press.

5 thoughts on “Statistical Machine Translation Gets Real: A Profile of Language Weaver”

  1. I’ve read the article about the new SMT project. Although I’m rather sceptical about the “revolution” in machine translation, it will be interesting to see the development of this conception in the nearest future. Surely, the rules-based MT systems, like Systran, PROMT, Power Translator, or Translation Memory sistems (Trados, Deja-Vu) cannot provide 100 per cent accurate translations. But they have, from my point of view, a great advantage: I can control the translation by changing contents of the dictionary (and even some translation rules) or editing the fragments of the text in the Translation Memory database. I can edit the input text making them more neat and clear to the machine. As for the SMT systems, I cannot realize how a final user could control the quality of translation. For example, if I need to change the translation of some word (which is can be easily done in traditional MT systems), I have to find and match an amount of text in two languages where this word is used and translated properly?

Comments are closed.