Google Making Language Translation Phone

Page 2 - Seeking answers? Join the Tom's Guide community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

kartu

Distinguished
Mar 3, 2009
379
0
18,930
Translation of the typed text is sooo far from "ok". Voice recognition is sooo "if there is additional noise, existing algorithms will break". Yet mega-cool creator of the "Chrome OS" (web browser in BIOS thingy is available for years now, mind you) wants to be even mega-cooler and starts to solve 2 unsolved problems at a time, heck it's even 3, you not only need to develop sophisticated algorithms, but they also must be fast enough, to allow on the fly translatoin.

Sigh...
 

doc70

Distinguished
Jul 3, 2009
186
0
18,630
[citation][nom]grillz9909[/nom]So humans can finally understand each other. Would we need to have huge wars against another country, if the people of both countries can understand what is going on? I think it would be easier to just sit down and talk. That's just my opinion though, maybe it's best that we alienate ourselves from... ourselves?[/citation]
BTW, language barrier is a lame excuse to go to war, and it has never been the case. Wars are idiotic on their own, they don't need any "excuses".
 

r0x0r

Distinguished
May 9, 2006
364
0
18,930
[citation][nom]archange[/nom]I'm less than enthusiastic about this approach, although I'm down with technology. This is nothing else than an excuse for people to stop learning foreign languages.My first gripe is that machine translation more often than not contains tons of errors. Even when it gets right, something is still missing. You lose human touch, sense and soul.The next thing that worries me is that this is just another thing that would make your brain go numb. When you take away the stimuli, that's when your brain loses interest and there goes intelligence down the drain.Secondly, we are becoming nothing more than couch potatoes, with machines doing the working, the speaking, the entertainment and whatnot. Ultimately, machines will realize that we're actually the weak link and will start doing the "living" for us as well.[/citation]

You also lose the opportunity to learn about the culture of the country of whose language you are learning (though for some people, culture is the thing that yoghurt has).

Also, when you go overseas people are far more appreciative when you speak their language. You get better service, etc (not to mention far easier to pick up the local girls) ;)

 
G

Guest

Guest
OK on languages... I live in The Netherlands. A traditional trade country. Most people know how to speak English, German, and occasionally French. That being said, the majority of us Dutch adults feels most confident speaking our native language, Dutch. But we know how to communicate abroad.
Knowing the most spoken languages on this planet (English, Portugese, French and Mandarin) one should be able to cope quite well, especially if everyone around the globe learned these languages. Or we could use Google's new invention.

I for one am looking forward to it.
 

annymmo

Distinguished
Apr 7, 2009
145
0
18,630
Do you think a real-world Babel Fish could eventually take away the need for people to learn multiple languages?

No because they need to be debugged! Again and again.
Language is a slow-moving target, even if it's slow-moving. It's still moving. Not mentioning that current translations contain tons of errors.
OpenOffice has no default sophisticated grammar checker. Which shows the lack of incentives/interest in those basic things. There doesn't seem to be interest in having a system that can do word/sentence hierarchy-aided text translation. (That would detect words context/meanings because some words in other languages have other meanings. This means that he translation device has a instance of the whole conversation to work in. ) Or action-mapped text-translation. (Where translation is mapped to actions and those actions translated to the other language. And we need other nifty/sophisticated features in place. But they are too low in the stack to be interesting. (They're basic functions/sophisticated. Other words boring and difficult. So not much progress there.

Anyway, there is the Simon Listens url:
http://simon-listens.org/index.php?id=122&L=1
 
Status
Not open for further replies.