Slashdot Mirror
Computer Translator Ready for Testing in Iraq
cgibby98 wrote to mention a Wired News story about a battle-zone translation technology that may allow near real-time conversations between English and Arabic speakers. From the article: "Funded by Darpa, the system would allow troops to communicate in Arabic through a laptop computer equipped with voice recognition and translation software. Troops could speak in English and have their words instantly translated into Iraqi Arabic, 'spoken' by a computerized man's voice. The program also translates Arabic into English. Will it replace the need for an interpreter when you're having some sort of high-level conversation? Absolutely not. But it is absolutely to the point where it could be useful in some carefully chosen situations."
I've thought of an implementation like this for some time, only I was thinking of the added element of sampling the user's voice with phonemes in the language to be translated to and then averaging that sound sample with the computer so that you could hear it somewhat spoken in the user's voice. Eventually, this would be a simple headset that could be worn, and you'd talk into a microphone and have some speakers around your ears broadcast what you said in translated form. Those speakers could also be a sort of unidirectional microphone for picking up on the foreign language-speaker's voice and translating it back.
It'd be for one-to-one conversations, of course.
Unless we get to a point where we can separate individual voices in real time and then translate them and have the computer dynamically assign a digital voice to each of the translations so we don't get a jumbled conversation.
I managed something similar a year or so back, in an attempt to create a 'babelfish'. Of course the input/output had to be specified, and it had a very limited range of languages - certainly no universal translator but it did use all free software (as that's all I have).
h p
0) Input recording of English languagge
1) Voice recognition software (Sphinx) pipes output to
2) Script using online translator to convert between language
3) Festival stumbles out an imhuman gramatically-wrong rendition of the input.
It wasn't exactly in realtime, I just fed it recordings, for which it would then output an audio file in the other language. The worst step was the voice recognition, which didn't work great even when given the output of the voice syntethisier.
Sphinx http://cmusphinx.sourceforge.net/html/cmusphinx.p
Let's hope Douglas Adam's words re the universal translator, the Babel fish,that "Caused more and bloodier wars than anything else in history" is not prophetic.
"No one in the military would make life or death decisions based on a machine translation." That's pure CYA for the first time this device gets someone killed. If a life or death decision needs to be made and the only thing you have at hand is a machine translation, what are you going to use?
a rmy">James T. Fallows, "The U.S. military does everything in Iraq worse and slower than it could if it solved its language problems. It is unbelievable that American fighting ranks have so little help. Soon after Pearl Harbor the U.S. military launched major Japanese-language training institutes at universities and was screening draftees to find the most promising students. America has made no comparable effort to teach Arabic. Nearly three years after the invasion of Iraq the typical company of 150 or so U.S. soldiers gets by with one or two Arabic-speakers. T. X. Hammes says that U.S. forces and trainers in Iraq should have about 22,000 interpreters, but they have nowhere near that many. "
I don't know how representative of the state of the art they are, but I've been massively underwhelmed at Babelfish's ability to understand foreign-language text and by ViaVoice's ability to understand speech. I can't imagine the effect of layering machine translation errors on top of machine voice interpretation errors.
According to href="http://www.theatlantic.com/doc/200512/iraq-
Instead of doing the obvious thing--give soldiers training in Arabic and offer big bonuses for Arabic-speaking recruits--the U.S. does nothing for a couple of years and then tries to throw a cheap technical fix at the problem.
If we must throw gadgets at the problem, why not a satellite phone linked to a big building full of human Arabic/English simultaneous translators?
"How to Do Nothing," kids activities, back in print!
I will believe that iraqis have democracy when the iraqi policemen can arrest Americans for crimes and when they joil OPEC so as to get the highest price possible for their oil.
Until those things happen Iraq is simply a US colony.
evil is as evil does
You couldn't be more correct here. I speak 4 languages pretty fluently, Persian (my mother language), English (thanks to my parents for forcing me to learn it when I was a child), Turkish (having studied in the Turkish speaking North Cyprus) and French (Oui je parle francais aussi).
Of these, three are Indo-European languages (French, English and Persian). Grammatically, they all follow the same rules (more or less), and making sentences in them is pretty similar. The place of adjectives, verbs and nouns might change, but you are fairly confident that you will be dealing with the same structures, like nouns and verbs. Not so in Turkish which is a Turkic language and follows completely different sentence structure. In the Indo-European languages, you might use different prepositions for the same purpose, but you at least use prepositions. Not in Turkish, in which everything becomes a suffix and there are no prepositions (I have heard other languages like Finnish to be similar).
What is the key to be able to successfully speak different languages? Just as the parent mentioned, it is to be able to THINK in it. When learning a new language, if from time to time you find yourself making some sentences in your head in that foreign language, then you are on the correct path. It shows that you mind is willing to think in that way. Thinking in your own language and trying to translate it at run-time (using a little computer jargon), is a recipe for disaster. The best you can hope for is to be a mediocre speaker.
Adding to the difficulty of translation is the cultural weight that a language carries with itself. A Language, especially those like Persian and Turkish which are not as widespread as English for example, is a symbol of that country and stands for all that nation's aspirations, dreams and fears. You can't learn Persian without becoming familiar with some historical figures and legendary myths and Pagan beliefs that Iranians still commonly boast about. You can't learn Turkish without learning different kinds of Kebab, and drinking Ayran (A juice made from Yogurt), and of course Atatürk. You can't learn French without knowing what Café Créme is, without knowing the difference between a red from Bourgogne from one from Bordeaux, and of course without knowing the significance of the Bastille Day.
Can a machine deal with all these peculiarities of languages? Quite frankly, unless dramatic improvements are made in artificial intelligence, I would say the answer is a resounding NO. Translation can't be done by just following some rules and using a dictionary. That approach can at best become yet another Babelfish, and we know how useful that is. Knowing a language means being able to think in that language, and machines currently don't "think". And unless there are dramatic breakthroughs in AI, I say we are decades, if not centuries away from a useful machine translator.
I am able to fluently speak in these 4 languages, yet I become completely speechless when I am given the task of translating from one of them to the other. That's why I could never comprehend how the minds of those UN translators who translate important political speeches on the fly, work. Knowing a language certainly doesn't mean you can translate from/to it as well. Translation is an art. A machine can never be a good translator, for the same reason that a machine can never become an artistic painter, or a composer.
--