Nov 062021

Machine translation is hard. To accurately translate text from one language to another, context is essential.

Today, I tried a simple example: an attempt to translate two English sentences into may native Hungarian. The English text reads:

An alligator almost clipped his heels. He used an alligator clip to secure his pants.

See what I did here? Alligators and clips in different contexts. So let’s see how Google manages the translation:

Egy aligátor majdnem levágta a sarkát. Aligátorcsipesz segítségével rögzítette a nadrágját.

Translated verbatim back into English, this version says, “An alligator almost cut off his heels. With the help of an ‘alligatorclip’, he secured his pants.

I put ‘alligatorclip‘ into quotes because the word (“aligátorcsipesz“) does not exist in Hungarian. Google translated the phrase literally, and it failed.

How about Microsoft’s famed Bing translator?

Egy aligátor majdnem levágta a sarkát. Aligátor klipet használt, hogy biztosítsa a nadrágját.

The first sentence is the same, but the second is much worse: Bing fails to translate “clip” and uses the wrong translation of “secure” (here the intended meaning is fasten or tighten, as opposed to guarding from danger or making safe, which is what Bing’s Hungarian version means).

But then, I also tried the DeepL translator, advertising itself as the world’s most accurate translator. Their version:

Egy aligátor majdnem elkapta a sarkát. A nadrágját egy krokodilcsipesszel rögzítette.

And that’s. Just. Perfect. For the first sentence, the translator understood the intended meaning instead of literally translating “clip” using the wrong choice of verb. As for the second sentence, the translator was aware that an alligator clip is actually a “crocodile clip” in Hungarian and translated it correctly.

And it does make me seriously wonder. If machines are reaching the level of contextual understanding that allows this level of translation quality, how much time do we, humans, have left before we either launch the Butlerian Jihad to get rid of thinking machines for good, or accept becoming a footnote in the evolutionary history of consciousness and intelligence?

Speaking of footnotes, here’s a footnote of sorts: Google does know that an alligator clip is a pince crocodile in French or Krokodilklemme in German. Bing knows about Krokodilklemme but translates the phrase as clip d’alligator into French.

 Posted by at 5:51 pm
Aug 122017

Machine translation still leaves a lot to be desired.

I was watching a cute YouTube video this morning, about a tiny kitten kept warm by a chicken.

The title of the video was in Spanish. My Spanish being nonexistent (in fact, at first I thought the title was in Italian) I used Google Translate. When I used Google Translate most recently, I was tranlating something into Hungarian, so that was the default target language. And Google dutifully translated the sentence, “gallina cuida gatito del frió”, into “Sült csirke cica-ellátás”.

Which means, literally, “Fried chicken kitty-supply”.

Not sure how Google managed to produce this gem of a translation. It offers a reasonably decent English translation: “Hen cares cold kitty”. But the French (“soins chaton de poulet frit” – “kitten care by fried chicken”) and German (“gebratenes Huhn Kätzchen Pflege” – “fried chicken kitten care”) versions are just as atrocious. And the Russian version? “Fried Chicken уход за котенком”… Google didn’t even deign to translate the “Fried Chicken” part (but where did it come from in the first place, when I am translating from Spanish to Russian?) although the rest of the translation (“care for the kitten”) is acceptable.

As I said… machine translation still leaves a lot to be desired.

 Posted by at 9:47 am
Feb 272013

I’ve been reading a lot lately about Quebec’s recent language police fiasco, an overzealous Office québécois de la langue française cracking down on an Italian restaurant for its use of the non-French word “pasta” and other, similar terms on its menu. Of course I’ve been reading a lot about it lately; apparently, its news coverage exceeded by a factor of 60 (!) the coverage Quebec premier Pauline Marois received during her recent trip to drum up foreign investment in the province.

Yes, I could go on lamenting the superficiality of the news media these days, and I think I would be right. But I am thinking about Pastagate now for a different reason: I am wondering if I am the only one seeing strong parallels between a zealous police force guarding the integrity of a language and a zealous police force guarding the integrity of a religion.

At least officers of the language police do not come with canes.

 Posted by at 10:49 am
Jul 012010

Here’s a headline from Google News that illustrates just how difficult it is for a non-native speaker of English (or, for that matter, for many a native speaker!) to understand journalists:

ABC Online: Houston expects changes for diggers under Petraeus

OK, so if you watch the news at all, you’d know that Petraeus is the US general who’ll be taking over in Afghanistan. But unless you also know that “ABC” can refer to the Australian Broadcasting Corporation, that Angus Houston is Australia’s Chief of the Defense Force, and that “digger” is a slang term for Australian or New Zealand soldiers, you could be excused if you thought that this was not an article title but a cryptic crossword entry.

 Posted by at 8:12 pm