Translation

[Kerim Friedman]

First off, a link to the story:

http://tinyurl.com/yy9u4n

Second, the key phrase that explains what the technology involved is:

"Using statistical probability tables built into their software, they
sort through likely word combinations to form phrases and full
sentences."

The use of statistical methods as opposed to rule-based methods has
been a major breakthrough in language translation, and has produced
significant improvements. The gains are most visible in precisely
those languages which are most different from our own and are
therefore hardest to model with rules.

You can test these technologies yourself using Google's language tools:

http://www.google.com/language_tools

Arabic and Chinese are listed as "beta" and are based on the same idea
as what is used by IBM in the news story.

Cheers,

kerim



On 12/21/06, Robert Lawless <robert.lawless at wichita.edu> wrote:
> Today's USA TODAY carries a story headlined "Military tests portable
> translators" about the effort to build a machine to do real-time
> translation relying on speech recognition software. The story relates,
> "This device has been particularly elusive when it comes to translating
> between English and complex languages such as arabic and Mandarin Chinese."
> The "chief technical officer for IBM Research" is quoted as saying, "You're
> talking about (languages) where everything is different – the words, the
> structure, even spelling and punctuation."
>
> Later in the story we learn that translation "is made more difficult by the
> great variety of spoken dialects and by large structural differences
> between English and arabic and Chinese. Chinese nouns, for instance, have
> no gender. In arabic, which does not use capital letters, many proper names
> are also common nouns or descriptive phrases."
>
> Huh? What am I missing here?
>
>
>
>