Online Language Identifier

Radiohead will be pleased to know that Xerox is at it again!  For
those of you who don't check Langmaker.com every two hours,
a resource was just posted about an online language identifier.
It can be found here:

http://www.xrce.xerox.com/competencies/content-analysis/tools/guesser-
ISO-8859-1.en.html

Basically it identifies the language that you put into the text
field (a sentence of five words or more).  It was reviewed on the
blog Tenser Said the Tensor.  The author put in Klingon, Quenya
and Sindarin.  Klingon apparently was fairly consistently identified
as Maltese.  I tried a couple of mine.  The results:

-Zhyler: Turkish.  (Right on!  That's what it's modeled after.  The
y with a diaresis and the presence of the letter "x" didn't seem to
phase it at all.)

-Kamakawi: Estonian.  (WHOA!!!  Way off!  This languages is
almost embarrassingly reminiscent of Hawaiian.  Now I'm
interested in finding out what Estonian is like, though...)

-Epiq: Latvian.  (Very interesting.  This is the language that
was inspired by Inuit.)

-Kelenala: Malay.  (Not bad, actually.)

-Njaama: Hungarian.  (Way, way off.)

-Sheli: Also Turkish.  (That one throws me for a loop.)

Anyway, try it out!  It's great fun!  Plus, this might help out the
"What language is this song/text in?" threads.  I hear for real
languages it's pretty accurate.

(Oh, a side-note: It has a fixed number of languages [46] it's guessing
from, and it lists them for you.  This list includes Esperanto, but
does not include any Austronesian language, I think...  [What is
Malay?]  That's probably why a language that looks like Hawaiian
looked so alien to it.)

-David
*******************************************************************
"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison

http://dedalvs.free.fr/

Replies