Google Translate Adds Nine More Languages
September 2nd, 2009 — Bruce Etling
While there is still a lot of English content on the Web, the percentage is shrinking fast, and those that want to understand what the rest of the world is talking about online have a potentially powerful new tool on their hands with Google translate, which has just added nine more languages: Afrikaans, Belarusian, Icelandic, Irish, Macedonian, Malay, Swahili, Welsh and Yiddish.
Icelandic? Yiddish? Hugh?
It turns out that Google chooses languages based on the amount of content available in those languages, not the number of speakers of a language or foreign policy concerns. As the Google Research blog says:
We’ve found that one of the most important factors in adding new languages to our system is the ability to find large amounts of translated documents from which our system automatically learns how to translate. As a result, the set of languages that we’ve been able to develop is more closely tied to the size of the web presence of a language and less to the number of speakers of the language.
Still, I’m excited by the project and the promised improvement in translation capacity over time, if that turns out to be true, and was quite happy when they added a beta version of Persian to the mix of languages earlier this summer (maybe Google isn’t totally immune foreign policy considerations after all).
As someone who has invested a lot of time trying to learn a foreign language or two (Russian has been my primary war of attrition), I always felt comfort in the fact that machines will never be able to translate very well. The results from most machine translators are often more humorous than useful. So, I was fairly impressed at how fast Google translate churns out a translation, if not necessarily with the quality of the end product. While the translation is better than most machine translators, it’s still not good enough to use free of a basic understanding of the language you are translating from if you want anything more than a general sense of an article or blog post (at least in Russian – I understand the quality varies among different languages). Humans (at least for now) are still better than computers at some things – but I applaud Google’s efforts so far on this one.
Click Here
September 2nd, 2009 at 3:21 pm
What’s fascinating here is the idea that their “system automatically learns how to translate” off these web docs – the idea that it is “learning” the language from the documents, improving over time etc. How one comes up with a “system” like that is beyond me – maybe they’ve coded Chompsky’s “universal grammar” …
September 22nd, 2009 at 4:39 am
Agree with you Karina. and google must do a lot of things to updating and make sure that Google Translate work for others languages
September 23rd, 2009 at 9:15 pm
Regardless of the number of the supported languages, machine translation can never guarantee accuracy of the translated text.