Have you ever wondered how the
Universal Translator works? I know I have. It always seemed a bit convenient to me that the UT not only can translate almost any language they come across, instantaneously, but it also knows when
not to translate something, like a phrase a character mutters in Klingon. And it does all the translating while also making it appear that your lips are speaking English...
One of the things I really enjoyed in Enterprise, even if it wasn't always done well, was the exploration of how the Universal Translator was developed and what happened when it didn't work. And I thought the character of
Hoshi Sato, as a kind of human Universal Translator was really interesting. It's a common enough trope in fiction that it must have some basis in truth that there are people who can pick up new languages after hearing just a little of them. Being the kind of person who's studied all sorts of languages and can't remember enough of any of them to even get by conversationally, I find this idea fascinating.
Anyway, on to
a little story in The Times. Apparently, Google is trying to create a Universal Translator of sorts. It wouldn't be quite as robust as the Universal Translator on Star Trek because it would be limited to languages in the database (52 at the moment), but it would combine
Google Translate with "high-accuracy voice recognition" to create software that allows the user to speak in one language and hear the translation in another language.
Anyone who has used either Google Translate or voice-recognition software knows that the result is likely to be a bunch of gobbledy-gook that may or may not come anywhere near resembling an accurate translation, but it will be a step closer to something resembling the Universal Translator. Now we just need to figure out how to keep it from translating
petaQ.
.
