Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

DARPA Starts Ultimate Language Translation Project

CmdrTaco posted more than 7 years ago | from the teach-it-klingon-first-please dept.

123

An anonymous reader writes "Defense Advanced Research Projects Agency (DARPA) has launched the ultimate speech translation engine project that would be capable of real-time interpretation of television and radio programs as well as printed or online textual information in order to be summarized, abstracted, and presented to human analysts emphasizing points of particular interest." If combined with the tower of babel project we discussed earlier, it could only lead to awesomeness.

cancel ×

123 comments

Sorry! There are no comments related to the filter you selected.

Wow that would be handy (1)

2.7182 (819680) | more than 7 years ago | (#16769305)

to understand DARPAese. (Try reading some of their PPT slides.)

Re:Wow that would be handy (1)

Zarniwoop_Editor (791568) | more than 7 years ago | (#16769375)

It certainly would be handy. I suspect that as even skilled human translators can mess things up at times that expecting any kind of automated system to produce good results will be an accomplishment indeed!

Re:Wow that would be handy (4, Interesting)

2.7182 (819680) | more than 7 years ago | (#16769571)

Seriously though, I just don't believe it. I've worked on a number of DARPA robot projects, and have heard a lot of their babble. They claim to be funding all these fantastic ideas, but none of them ever work except in a limited capacity. The robot projects I worked on were very lame in that DARPA created these really specific environments for the robots that were light years away from what they were saying they were really going to do. All of the Universities involved failed to accomplish even the simplest tasks. So my experiance with them is that they talk a big talk, and no one ever goes back to check "hey did you really ever do that ?" Now granted, some of their work is supposed to be high risk, but they never emphasize which projects are expected to have a high failure rate. Largely because they don't care. It's really all about funding your academic buddies or whoever is going to be able to scratch you back in some way. It is very much an old boys network, with an emphasise on PR and not much about real science. Much like the MIT media lab. (Just thought I'd get another jab in there....)

Re:Wow that would be handy (1)

jotok (728554) | more than 7 years ago | (#16769683)

They claim to be funding all these fantastic ideas, but none of them ever work except in a limited capacity.

Like that internet boondoggle?

I keed, I keed :)

Re:Wow that would be handy (1)

2.7182 (819680) | more than 7 years ago | (#16769735)

That was ARPA, not DARPA. And I have little doubt that the culture of ARPA in the 60's is different then the culture of DARPA now.

Re:Wow that would be handy (1)

hkgroove (791170) | more than 7 years ago | (#16769861)

Same group, slightly different name over the years. Just like Prince.

Seriously though, I just don't believe it. (3, Insightful)

Anonymous Coward | more than 7 years ago | (#16770153)

The US department of Defense is openly claiming to be able to solve one of the world's hardest AI problems, and you don't believe it? Big surprise.

If the US military had anything close to real A.I., you wouldn't hear about it. It would be a classified information.

The NSA would love to have anything close to a system capable of understanding language as well as a native speaker can; as would the CIA, or any other clandestine organization. Any system smart enough to understand and generate English probably also came with a breakthrough in CS theory that will give them better tanks, planes, and communications systems. And those would be classified, too.

In short, this is just an excuse to spend money, and to hide the funding for any secret research projects that they really are working on.

Actually... (3, Interesting)

technococcus (990913) | more than 7 years ago | (#16771069)

I attend one of the many Universities where DOD research is currently being conducted. Portions of our graduate student body and faculty are working on the powered armor concept in conjunction with UC-Berkley (they're doing the frame and kinematics, we're doing the control theory/system and power supply). We're actually making quite a bit of progress in the field of alternative batteries (the current iteration is a peroxide-fueled hydraulic hybrid-type system widget) and mechanical interface control theory application. So, while God knows we won't see cap' troopers in 'suits any time soon, we are at least progressing towards that end while developing widely applicable technologies along the way (this is, if I may remind you, the way many technologies we love dearly were spun off from the space program et. al).

You have to walk before you can run (4, Insightful)

Zontar_Thing_From_Ve (949321) | more than 7 years ago | (#16771341)

Seriously though, I just don't believe it. I've worked on a number of DARPA robot projects, and have heard a lot of their babble. They claim to be funding all these fantastic ideas, but none of them ever work except in a limited capacity.

This is a big pipe dream that is extremely unlikely to work any time soon. How do I know that? Right now, I think it would be reasonable to conclude that computer technology today is good enough to do accurate text translation. Can it? Well, it depends on how picky you are. There are always mistakes, sometimes glaring ones, in text to text translation programs. I can speak Russian and for convenience (to get quick rough translations) at one time I owned what is probably the best Russian-English text translation program. It's much more accurate than Babelfish. It still left a lot to be desired. It would be about 80-90% accurate, but no more. I remember one time when it took a statement in Russian that said "I absolutely would not mind to tell you about ..." and translated it as "I absolutely would mind to tell you about ..." which is the exact opposite. Many languages, such as Russian, Spanish and Portuguese (and no doubt others) use double negatives to express negation. "I don't know nobody" is quite correct in Russian, Spanish and Portuguese although it is quite grammatically incorrect in English if your intention was to say "I don't know anybody". Programs that translate into English from languages that use double negatives often fail to correctly translate the negation. Maybe there are some that get it right, but I've never seen any. Text translation programs are very poor at distinguishing between words that have uses as different parts of speech. Here's an example:

She sings like an angel.

In this sentence, "like" is an adverb, but it can also be a verb ("She likes to go shopping."). A text translation program might fail to correctly understand that "like" is an adverb here and say something like:

She sings and angel is pleasing to her.

I could give a lot more examples, but these are enough. If we can't even do a better job right now at text translation, how on earth is DARPA going to get speech translation right? This is the kind of project that gets funded by idiots who have never studied foreign languages and believe that the Star Trek idea of a Universal Translator is only a few years away.

Re:You have to walk before you can run (1)

Venner (59051) | more than 7 years ago | (#16771703)

The time-honored CS example is:

Time flies like an arrow.
Fruit flies like a banana.

Always liked that one.

Re:Wow that would be handy (0)

Anonymous Coward | more than 7 years ago | (#16769761)

It would also be handy to be able to interpret idiots who post half their thought in a box labeled "Subject" and the other half in "Comment".

That is so rude, dude. Learn how to use a subject field. Learn how to speak in full sentences. geez. Slashdot is full of you idiots.

Really I hate it (1)

Feminist-Mom (816033) | more than 7 years ago | (#16769875)

when people do that.

But seriously, I've basically made my living off of DARPA grants and I fully support the criticism leveled at them above. It is truly a classic government buearacracy, very wasteful, not entirely straight about what they are doing, and you have to have personal connections to get money from them.

Re:Really I hate it (1)

Beyond_GoodandEvil (769135) | more than 7 years ago | (#16770545)

It is truly a classic government buearacracy, very wasteful, not entirely straight about what they are doing, and you have to have personal connections to get money from them.

In other words, just like 90% of the rest of the world. ;)

Re:Wow that would be handy (1)

andy666 (666062) | more than 7 years ago | (#16770049)

I wish I had mod points to mod the parent up. I have sat through some DARPA presentations that look absolutly crazy and incomprehensible. If there is anyone out there who doesn't know how to make good ppt slides, it is government agencies. Do I have any support from anyone out there ?

Oh wow...AI tanslators!!! Who would've thunk it? (0)

Anonymous Coward | more than 7 years ago | (#16771871)

Alright...anyone who is not aware that all these AI "translators" have been a complete failure should grab a AI history book. It might be incrementally better to what's out there; however, I'm almost certain it'll fail to accomplish it set out to do...at least from any practical perspective.

FP (-1, Flamebait)

Anonymous Coward | more than 7 years ago | (#16769307)

cock in your ass

Autobots, Transform! (1)

eldavojohn (898314) | more than 7 years ago | (#16769319)

Defense Advanced Research Projects Agency - DARPA - is working on the ultimate speech translation engine that would be capable of real-time interpretation of television and radio programs as well as printed or online textual information in order to be summarized, abstracted, and presented to human analysts emphasizing points of particular interest.
In unrelated news, a user named DARPABOT has made the Slashdot Hall of Fame [slashdot.org] under most active submitters at over 1000 in under a few weeks time, crushing prostoalex.

If combined with the tower of babel project we discussed earlier, it could only lead to awesomeness.
By 'lead to awesomeness' do you mean 'lead to you not having to attempt to edit summaries and fail at both grammar and spelling'?

Take us to your leaders... (1)

Noryungi (70322) | more than 7 years ago | (#16769327)

... Now in 54 flavours! :-)

Awesome? WTF?? (4, Insightful)

BadAnalogyGuy (945258) | more than 7 years ago | (#16769333)

If you consider that now the government will be able to spy on you in your native language to be awesome, then I suppose giving the Feds this sort of technology can only lead to awesomeness.

Surveillance of civilian populations under the guise of "monitoring terrorists" is not something that I'd consider awesome. Irksome, yes. But not awesome.

Re:Awesome? WTF?? (4, Interesting)

CRCulver (715279) | more than 7 years ago | (#16769433)

As if the government doesn't already have legions of translators at the ready. Military linguists are trained at Defense Language Institute [dliflc.edu] at the Presidio of Monterey. I studied Chinese there while serving in the Navy, and while most of my fellow enlisted servicemen were likewise studying languages of some clear strategic value, there are also courses in various other languages for officer exchange programs, as well as the occasional course in something really exotic. Combined with the simple possibility of the government paying a native speaker to work for them, this means that the government already has the language skills it needs even without a whizbang translation machine.

Re:Awesome? WTF?? (1)

b0s0z0ku (752509) | more than 7 years ago | (#16769589)

As if the government doesn't already have legions of translators at the ready.

Assuming that this system can recognize voice well, and then convert it into text in preparation for translation, this is already saying a lot. This means that phone conversations can in theory be automatically logged as text, which requires much less storage space than audio.

-b.

Re:Awesome? WTF?? (1)

TechnoBunny (991156) | more than 7 years ago | (#16769813)

Wouldnt the original audio need to be stored as well, for evidential reasons?

Re:Awesome? WTF?? (1)

b0s0z0ku (752509) | more than 7 years ago | (#16769893)

Wouldnt the original audio need to be stored as well, for evidential reasons?

Depends what you want to do with it, and assuming that our court system is intact and more or less unchanged in 20 years. Besides, there's always the option of kidnapping and "disappearing" miscreants. I'd hate to see what would happen, with the full consent of the majority of the lumpenproleteriat, if another 9/11-scale (or worse) terrorist attack occurred on US soil.

-b.

Re:Awesome? WTF?? (1)

DragonWriter (970822) | more than 7 years ago | (#16770019)

Wouldnt the original audio need to be stored as well, for evidential reasons?


The Department of Defense isn't particularly interested in evidence. Indeed, in many cases once they have the information they need to make a decision and the decision is made, it seems they'd be happier if the underlying original data was irretrievably lost to prevent any after-the-fact criticism of either their decisions or their methods.

Re:Awesome? WTF?? (0)

Anonymous Coward | more than 7 years ago | (#16769971)

> This means that phone conversations can in theory
> be automatically logged as text

Last I checked the NSA already has filed patents a couple years ago for just such a thing. IIRC the patents covered a text-to-speech method and subsequent Google-style search of the transcribed conversations. This may already be in full world-wide use via Echelon etc..

Re:Awesome? WTF?? (1)

cyberon22 (456844) | more than 7 years ago | (#16770305)

Considering that I've personally handled documents that the US embassy has "outsourced" for Chinese-English translation in Beijing, I think your confidence that the US has enough skilled translators in-house is grossly misplaced. Chinese is my second language and I am good at it, but I am not even an American citizen. And although I can't speak for military training, I have met people who have been trained by the State Department and found that very few of them have pushed beyond middling Chinese despite having serious advantages in time and funding for training.

That being said, let me praise as exceptions those people I've met who trained at IUP when it was in Taiwan, and less so in Beijing. With those few exceptions, everyone I've met who is remotely decent at Chinese has spent considerable time in China and mops the floor with those who have studied abroad.

As far as the tech goes, I'd respectfully suggest you're wrong on the need for more and better tools. The problem is that translation is really a small and limited domain for the use of bilingual NLP systems. I run an educational project which is doing something the private sector simply is not: developing Chinese-English machine annotation and translation technology with a focus on providing educational annotations for students. The technology itself is quite cool and if you are still studying Chinese you should actually check it out [adsotrans.com] . We are currently the only place on the web where you can get word-by-word explanations of everything from newspapers [newsinchinese.com] to excepts from classic novels like Dream of the Red Chamber [adsotate.com] .

Getting back to the point.... the market that is emerging for this sort of technology is happening in places like enterprise search, contextual analysis of Chinese documents for search and other areas where there is a need for massive data analysis and you simply cannot rely on human translators. The real tragedy is that the market is not here yet, and the money is all going into closed research programs like those cited in the article. Seeing the US government simply stuffing millions and millions of dollars into closed research consortiums does not help what we are doing one bit. Nor does the status quo is a situation through which NIST holds "open" machine translation competitions where the results are not even made available for public comparison. More's the pity....

human vs machine translation (0)

Anonymous Coward | more than 7 years ago | (#16770357)

that may be true for most languages, but just about every government defense/security agency i've heard of is desperate for speakers of arabic, persian, malay, javanese, pashtu, etc. And they much prefer native speakers, for obvious reasons -- very few people ever achieve native-like proficiency in a language they learn after 12 or so (the critical period for language acquisition is pretty brief, sadly).

that said, a skilled non-native translator will no doubt beat the crap out of a computer. this DARPA project sounds pretty far-fetched, even for them. i doubt they'll end up with much more than the rudimentary string of words altavista creates. (seriously, if you want a good laugh, hook babelfish up to a chinese or japanese newspaper -- see if you can guess what the headlines are actually about).

Defense Language Institute (1)

Kozar_The_Malignant (738483) | more than 7 years ago | (#16770467)

>the government already has the language skills it needs even without a whizbang translation machine.

Sadly, they don't. The FBI has something like two guys who speak Arabic, and there are numerous instances in the news recently where some fed is bewailing the lack of language skills in his department. On a diplomatic note, how many US Ambassadors [state.gov] actually speak the language of their host country? It might be useful if they had some way to understand the locals.

Re:Defense Language Institute (1)

zebul0n (84908) | more than 7 years ago | (#16770929)

Why would they need to understand them when they b0mb them?

Primary language (0)

Anonymous Coward | more than 7 years ago | (#16772215)

most of my fellow enlisted servicemen were likewise studying languages of some clear strategic value

Now if we could just teach English to the Marines.

Re:Awesome? WTF?? (0)

Anonymous Coward | more than 7 years ago | (#16769549)

If you consider that now the government will be able to spy on you in your native language to be awesome, then I suppose giving the Feds this sort of technology can only lead to awesomeness.
Uh, about 99% of the people to whom this is "the" government speak English as their native language. So, uh, it's not really changing anything, even if it works, which it won't, because machine translation is about as real as Star Trek.

Just sayin'.

I read that as... (1)

rHBa (976986) | more than 7 years ago | (#16769659)

I read that as "Defense Against Research Projects Agency"!

Re:Awesome? WTF?? (-1, Offtopic)

Milwaukee (716971) | more than 7 years ago | (#16769709)

Didn't you read the headlines this morning? Nancy Pelosi is in power now. The government will now bring us nothing but sunshine & good will.

Don't be too afraid... (1)

Olivier Galibert (774) | more than 7 years ago | (#16771285)

... ASR on people talking to each other naturally Just Doesn't Work[tm]. As in 70-80% error rate or worse.

Gale is about TV/Radio news, not random people conversations.

    OG.

Re:Awesome? WTF?? (1)

4D6963 (933028) | more than 7 years ago | (#16772167)

now the government will be able to spy on you in your native language

I can imagine your typical terrorist conversation translated using this system :

-After this the friends when it jumps the operation?
-Not white I go seeing
-Into the correspondence, and to know, you have?
-I am caused
-And on the other hand are their blond as?
-It goes, or
-Or he has
-The God is large!
-The God is large!

Re:Awesome? WTF?? This... could... (1)

davidsyes (765062) | more than 7 years ago | (#16772713)

be very suck. If you anything secret to say you better be hurry.

Ultimate Defense (4, Funny)

Speare (84249) | more than 7 years ago | (#16769391)

Just feed this new system a few reruns of Japanese television game shows. After that, we will be safe from automated snooping for at least another decade. As a plus, all artificial intelligence projects at the DARPA will be set back by another decade as well.

Humans??? (2, Insightful)

Anonymous Coward | more than 7 years ago | (#16769409)

What's wrong with using humans? This is exactly what humans are good at. While there most certainly are fields where machines can replace humans, this is _not_ one of them.

http://lyricslist.com/lyrics/artist_albums/16/ac-d c.php/ [lyricslist.com]

Re:Humans??? (2, Funny)

3waygeek (58990) | more than 7 years ago | (#16769573)

Yes, but Hoshi Sato won't be born for another 100 years or so...

Re:Humans??? (4, Interesting)

Protonk (599901) | more than 7 years ago | (#16769605)

One of the problems with using humans is that they are expensive--the other is that they become bored easily. It isn't like the defense establishment isn't using human translators, the NSA is the largest employer of translators [amazon.com] in the world. They use humans in every listening post out there, but for the same reasons that humans make lousy airport security sceeners, they make poor translators AND intelligence analylists. This isn't saying that machine translators are a panacea, but they can solve a small section of the problem that we have been trying to solve with a very human capital intensive solution for years now.

Re:Humans??? (1)

Scott7477 (785439) | more than 7 years ago | (#16770731)

I lived in Japan for two years and earned a minor in the language at a major private university in the US. Toward the end of my time in Japan, my skill with the language was sufficient that when I spoke to native Japanese on the phone, frequently they thought I was actually Japanese rather than a foreigner. I seriously considered working for the US government as a way of exploiting my Japanese skills, but I concluded that spending my days translating the kind of documents that the US government would be interested in would be incredibly, horrifically boring.

Re:Humans??? (1)

JonTurner (178845) | more than 7 years ago | (#16769681)

>>What's wrong with using humans?

They're slow, and scarce and don't work 24-7. *If* the software has progressed to the point that it's "good enough" (that's a big IF) then a massive farm of machines could simultaneously monitor all communications (tv, email, phone, IM, etc.), summarize, and filter out anything interesting, looking for trends. Think Really Big Brother.

Lots of reasons (4, Insightful)

phorm (591458) | more than 7 years ago | (#16769983)

To properly translate all the nuances of some languages actually requires a lot of skill, and sometimes translating can be ask much interpreting as anything. Granted, this is something a human could handle better than a machine, but the problem is that humans also have a bias. Yes, there have been cases wherein human translation has caused problems because of bias or even due to being outright wrong.

I reminds me of the old joke:

Guard: Now tell me where you hid the money, or you will suffer
Translator: Tell him where the money is, or you will suffer
Prisoner: I'll never speak
Translator: He says he won't tell you
Guard: *putting gun to prisoner's head* Tell him I will blow his brains out if he doesn't tell me immediately
Translator: He will shoot you in the head unless you tell him now
Prisoner: I buried a million dollars under the floorboards in the old woodshed
Translator: *pauses* He says you don't have the guts to shoot him...

Re:Humans??? (2, Insightful)

DragonWriter (970822) | more than 7 years ago | (#16770183)

What's wrong with using humans?


The number of humans that the Pentagon can afford to employ with adequate skill in the languages it wants to target are inadequate to process all the channels of information it would like to filter for potentially interesting information, further, the more humans know what information is being looked for (and what is flagged), the greater the security risk.

Three words. (1)

Rob T Firefly (844560) | more than 7 years ago | (#16770465)

All your base. [wikipedia.org]

not enough of them (1)

peter303 (12292) | more than 7 years ago | (#16771671)

We could probably collect as many cellphone and internet messages as we want, but there arent enough people to sift through them.

Several terrorists in Colorado Supermax prisons sent over a hundred unread Arabic letters overseas because they have just one part time guy reading them down there. Quite a scandal there.

Cool (1)

thejrwr (1024073) | more than 7 years ago | (#16769431)

Dragon Speak is going out of business of guess

Re:Cool (1)

aicrules (819392) | more than 7 years ago | (#16770043)

Naturally!

awesomeness, in terms of megatons (1)

JonTurner (178845) | more than 7 years ago | (#16769441)

awesomeness, to be sure, if you consider the ultimate outcome. Remember, this is DARPA, so they're looking at potential military applications. I read it as: "translate (military) communications in real-time, ... then destroy one or both parties."

Re:awesomeness, in terms of megatons (1)

Milwaukee (716971) | more than 7 years ago | (#16769595)

DARPA research & innovation has had many positive impacts on our lives. The most obvious example is the internet. Presuming it is successful, this new translation technology would definitely shrink the globe again, bringing us all closer. Isn't that a good thing?

Re:awesomeness, in terms of megatons (1)

the eric conspiracy (20178) | more than 7 years ago | (#16769663)

this new translation technology would definitely shrink the globe again, bringing us all closer. Isn't that a good thing?

I don't what to know what people really think of me.

Re:awesomeness, in terms of megatons (0)

Anonymous Coward | more than 7 years ago | (#16769885)

...bringing us all closer. Isn't that a good thing?


No, it isn't; it's fscking terrible. More and more people are moving closer and closer to my cave every year!

When will it affect me? (2, Interesting)

jandrese (485) | more than 7 years ago | (#16769479)

All I really want is a free online translator for web pages (ala Babelfish and Google) that aren't terrible at it. Seriously, the quality of Babelfish translations has stayed constant since it came on the scene in the late 90s, even though machine translation in general has made some rather significant advances. I don't really use them enough to justify plopping down $500 on the professional packages, but the current systems are just terrible.

Re:When will it affect me? (2, Funny)

Creepy Crawler (680178) | more than 7 years ago | (#16769815)

You get what you paid for.

w0t j00 s4y (0, Offtopic)

Anonymous Coward | more than 7 years ago | (#16769519)

w0nT b3 phUn 1ph th3y d0nT d3w l337sp3ak

owned [infiltrated.net]

"My hovercraft is full of eels." (0)

Anonymous Coward | more than 7 years ago | (#16770329)

"My hovercraft is full of eels."

By the way, mod parent up. "Offtopic" is not a good replacement for "I don't get the joke".

Re:"My hovercraft is full of eels." (1)

Jinky Williams (975076) | more than 7 years ago | (#16770945)

Agreed.

i /\/\34|\| i (4|\| |)0 t3|-| |-|4|2|)(o|23 1337 7|-|47 |-|4$|\|7 833|\| |_|$3|) $!|\|(3 \/\/4|23Z |)4'/$, 4|\||) \/\/|-|!13 i7$ |\|07 |)1|=|=!(|_||_7 70 7|24|\|$|_473 !7 |_4|293|_'/ |_|\|!|\|73|_|_!9!8|_3 70 |\/|0$7 |>30|>|_3 4|\||) 7|-|3|23|=o|23 4 900|) $|_|8_|3(7 |=0|2 7|24|\|$|_47!0|\|

May I be the first to say... (0)

Anonymous Coward | more than 7 years ago | (#16769537)

Es una trampa!

can't help it (1)

Programmer_In_Traini (566499) | more than 7 years ago | (#16769545)

its lame, its old and yet i cannot help it

i can see a translated japanese movie coming...

"all your base are belong to us, make your time"

Re:can't help it (-1, Offtopic)

Anonymous Coward | more than 7 years ago | (#16769657)

it's lame, it's old and yet you still got it wrong...

you fail it

Babelfish: "There is a chisel in my dog." Me: WTF? (1)

zooblethorpe (686757) | more than 7 years ago | (#16772727)

Seriously, try it. Input the sentence, "My dog has fleas." Go from English to Japanese, copy and paste the Japanese into the entry box, and translate back to English. "There is a chisel in my dog."

Just one of many reasons that I'm not that worried about my career as a Japanese - English translator. :P

My Criteria For Success (0)

Anonymous Coward | more than 7 years ago | (#16769583)

As long as it can handle swear words and a good variety of sex acts I'm interested in I say thumbs up!

Too much too soon, or tackling wrong problem? (5, Interesting)

Salvance (1014001) | more than 7 years ago | (#16769607)

This project, along with CMU's Tower of Babel, certainly get props in the coolness category, but the practicality is still lacking. I believe DARPA is barking up the wrong tree for now, or at least biting off more than they can chew.

Speech Recognition is the hardest problem to tackle on the path to recognition, and MUST be addressed before there is a viable real-time (or even delayed) translation engine. Currently, even the best speech recognition software can achieve at best ~80% accuracy when faced with a large vocabulary with no limits on speakers/dialects, and this level of accuracy is typically not achieved in real-time. While this 80% level is actually pretty good when transcribing to text (since the reader can typically decipher what the computer meant), it's downright awful if trying to translate the resulting text to another language.

For example, if I say "I like ice cream" into voice recognition software and 'hears' "I like, I scream", the reader might understand what this means, particularly if they say it in context and aloud. However, let's say we translate each sentence into Spanish ("Tengo gusto del helado" and "Tengo gusto, yo grito" respectively, according to Babel Fish), and the speaker would be completely lost as the out of context phrases don't sound anything alike. In a natural language translation, even under relatively accurate recognition scenarios, would be frought with misunderstandings.

Once speech recognition is tackled, it's just a matter of translation then voice synthesis. Fortunately these problems aren't nearly as difficult, and current solutions would suffice (with the only pitfall being poor grammer in the destination language, and a robotic sounding voice).

Re:Too much too soon, or tackling wrong problem? (1)

Rocketship Underpant (804162) | more than 7 years ago | (#16769931)

"Fortunately these problems aren't nearly as difficult, and current solutions would suffice (with the only pitfall being poor grammer in the destination language, and a robotic sounding voice)."

*Good* translation is extremely difficult unless you stick to "see Dick run"-type sentences. Good translation between non-related languages (like Japanese and any Indo-European language) doubly so.

Advanced machine translation on par with human will require nothing less than artificial intelligence, most likely.

Re:Too much too soon, or tackling wrong problem? (0)

Anonymous Coward | more than 7 years ago | (#16769991)

State of the art commercial speech recognition such as Dragon Naturally Speaking 9 is better than you suggest - accuracy percentage is in the high 90's, and it has sufficient understanding of English to resolve phonetic ambiguities... it's not relying on words sounding different to pick the right one. Huge vocabulary, continuous speach, and close to real-time too (a second or so lag).

Re:Too much too soon, or tackling wrong problem? (1)

Red Flayer (890720) | more than 7 years ago | (#16770105)

"I believe DARPA is barking up the wrong tree for now, or at least biting off more than they can chew."

I think what you're trying to say is that DARPA isn't capable of developing speech recognition software equal to the task of real-time translation.

I'm sure that DARPA is fully aware that the biggest block to real-time translation is speech recognition -- that's why they are funding this project -- because it is (1) beyond the scope of what private enterprise is currently capable of without cash influx and (2) because it would be extremely useful to have such a tool, and to have the IP rights to it.

Speech recognition is a part of translation, not outside it.

Re:Too much too soon, or tackling wrong problem? (1)

cyberon22 (456844) | more than 7 years ago | (#16770503)

Not sure about Arabic which is probably where the moeny is right now, but speech recognition is much less of a problem in languages like Chinese than in romance languages like English which conjugation, inflection and the sort.

On the other hand, text translation is much harder in Chinese than in romance languages, in large part because of the lack of conjugation, etc.

Re:Too much too soon, or tackling wrong problem? (0)

Anonymous Coward | more than 7 years ago | (#16770673)

Your 80% estimate is innacurate. Please cite your source.
I am a developer for a LVCS engine which is more towards the 90% accuracy mark.
For a concrete example, the CMU sphinx 3.5 decoder can be tuned to achieve ~8% WER (word error rate) at about 3xRT.

Translation is actually the hard part in this system. MT (machine translation) is a difficult task, but _a lot_ of money is being put into it right now (especially by Google). There should be some significant breakthroughs in the next decade. Projects like this one only help spread the academic knowledge and allow companies to implement concrete solutions for these broad tasks.

Re:Too much too soon, or tackling wrong problem? (1)

Olivier Galibert (774) | more than 7 years ago | (#16771215)

It's broadcast news in arabic and mandarin. I don't have the latest numbers handy but 80%, well, 20-25% error rate, nobody in the ASR community uses accuracy, is pretty much correct for arabic. I think mandarin is better though. Arabic has some unique problems, the main one being that there is no such thing as arabic in the first place.

    OG.

Re:Too much too soon, or tackling wrong problem? (1)

maxume (22995) | more than 7 years ago | (#16770773)

DARPA just wants a way to figure out what documents are most interesting to give to a human to translate. I don't think the 80% accuracy is going to be all that huge a problem for something like that.

Re:Too much too soon, or tackling wrong problem? (1)

Bazouel (105242) | more than 7 years ago | (#16770799)

They should also start by making a translation engine that actually works...

See, even for the most simple translation of "I like" in Spanish, Babelfish is wrong. The good translation is "Me gusto", not "Tengo gusto" which means "I have taste".

I am trying to learn Cantonese and you have no idea just how "stupid" the current translators are...

Re:Too much too soon, or tackling wrong problem? (1)

xtracto (837672) | more than 7 years ago | (#16771287)

And to make things worse, "I like" does not means "Me gusto" but "Me gusta" as the first person conjucation of the verb.

So, what great parent wrote as "I like iceream" would be translated as "Me gusta el helado" and "I like, I scream" would be "Me gusta, yo grito".

I agree with GP about the speech recognition problem being one of the problems to cope before having a real-time translation tool. Some have said that the current technology (Dragon Speaking 9, etc) achieve 90% of accuaracy, but the issue is that the 90% is on ideal conditions, try the experiment of recording a Friends show (or simpsons or any other tv program) and try to introduce it with those programs. There is so much noise and variation in the speech that they cant make any sense.

Of course the translation problem is another difficult problem. Now my question is, isnt it possible to avoid the Speech to Text conversion in translation? wouldnt it be possible to translate directly from voice? it would mean translating a specific sound to a equivalent sound in another language which would mean having huge sound databases but hey, that might work and you will have to beat only one problem.

You don't know what you're talking about... (1)

Olivier Galibert (774) | more than 7 years ago | (#16771119)

Translation *is* the hardest part in the Gale project. So much harder than in the current evaluations the translations are so bad that the impact of the ASR errors on the final result is not significantly detectable it seems. We hope that the MT teams are going to make some massive progress fast (they may, they get a *lot* of new data from the project) so that working on the ASR actually means something, but more importantly so that the project goes on.

    OG.

Re:Too much too soon, or tackling wrong problem? (1)

avir (1020245) | more than 7 years ago | (#16772413)

I beg to disagree with you on the relative difficulty between speech-to-text (STT) and machine translation (MT). The state-of-the-art in broadcast news transcription is currently over 90% accurate - using 100 minus word error rate (WER) - in English and close to 90% in both Arabic and Chinese. Also, English conversational telephone speech transcription reached over 85% accuracy during the DARPA EARS program. However, translation accuracy - using 100 minus human-mediated translation error rate (HTER) which is the official metric in DARPA GALE - is only around 80% on both Arabic-to-English and Chinese-to-English.

To counter your last statement, experiments carried out before the GALE 2006 evaluation showed that the translation accuracy of STT output is pretty much the same as the translation accuracy of the STT reference transcripts. This is clearly due to the poor performance of the current state-of-the-art MT. Most of the research in GALE is currently tackling MT and only when the MT is good enough, the STT errors will begin to make a difference.

This could be dangerous... (2, Funny)

FuryG3 (113706) | more than 7 years ago | (#16769661)

Ultimate Language Translation researchers should probably compete along a stretch of the Mojave desert so as not to injure or offend nearby native speakers

Civilian use of such a thing (1)

cucucu (953756) | more than 7 years ago | (#16769685)

Obviously such a thing will not work well without an advertising filter (imagine an analyzer sifting through washing powder ads).

So they will have to develop one.

This will be integrated into VCRs to stop/start recording when advertising starts/stops.

Great!

Yeah... Billions of Dollars Later... (1)

Dark Leaper (989158) | more than 7 years ago | (#16769777)

Slashdot headline ten years from now... "Creators forgot to implement 1337 speak into translator matrix..."

Better yet... (0)

Anonymous Coward | more than 7 years ago | (#16770563)

Headline: "Lack of translator support kill "l337" and "IM" speak. World rejoices as collective intelligence level rises."

yeah.. awesomeness... (1)

Churla (936633) | more than 7 years ago | (#16769803)

Or the WRATH OF AN ANGRY GOD! heh

"Hey... didn't I make them all speak different languages to teach those uppity humans a lesson? Now they what? The end routed me on that one? Oh I don't think so!"

But will it translate... (1)

Billosaur (927319) | more than 7 years ago | (#16769921)

...Romulan, Klingon, and Vulcan?

They need to start working on... (1)

shirizaki (994008) | more than 7 years ago | (#16769923)

The metal gear project. I mean, honestly, the DARPA chief isn't going to be jailed in the secret arctic base for the ultimat language translator.

Interesting, (2, Insightful)

DragonWriter (970822) | more than 7 years ago | (#16769977)

Seems likely to be very useful for specifically what they suggest it is for (flagging potentially interesting material for further review by human analysts, a kind of time-saving filtering device for the limited pool of translators available.)

But beyond that, I wouldn't give too much faith in any kind of mechanical translation as particularly reliable on its own except on narrow kinds of material. It conceivably might work for strictly literal usages, or for fairly stable idiomatic uses, but unless you have frequent collection and incorporation of usage data from every culture and subculture that may be a source of translated material, its going to fail, sometimes subtly and sometimes spectacularly, for a lot of idiom. Similarly, even within the same language, different groups using it will have different idiomatic uses that sometimes will produce different or opposing meanings for similar usages, which will require accurate identification of the source at more than just the language level to get correct results from. There's a lot of evolving cultural context that informs the use of language...

exponential growth (1)

Silver Sloth (770927) | more than 7 years ago | (#16769981)

From TFA
As you can see all these projects are a far cry from what DARPA wants. But given time and money something more advanced would surely come out and eventually would be available for civilian use as well.
Well, err, yes, but, I have enough difficulty understanding Jordies and Glaswegans, and they're speaking the same language as me (nominally). Understanding 200 or so words when carefully spoken is a huge step from simultaneously interpreting random speech and I'm sure the problems will rise exponentially. File this one under the 'maybe, someday, but don't hold your breath' dept.

The REAL universal translator (1)

SethEaston (920552) | more than 7 years ago | (#16770109)

While being still quite far from an adaptive Star Trek-style universal translator, it is conceivable that one day with the help of portable devices (like Palm/cell phones/iPod) that we could indeed have an on-demand personal translator that would work anywhere. I think this is the beginning of such a capability.

screaming "failure" (1)

sohp (22984) | more than 7 years ago | (#16770193)

What is it about "ultimate, do-everything" project sponsored by the government that sets off every alarm bell signaling imminent failure?

Back in 2000 (0)

Anonymous Coward | more than 7 years ago | (#16770327)

Back in 2000 or 2001, I saw IBM television ads about telephones that translate in real-time. In the ad an English-speaking woman was speaking to a Turkish person, with the telephone doing the translation. This was promised in the "near future" by IBM. Anyone know what became of it?

Maybe not! (0)

Anonymous Coward | more than 7 years ago | (#16770761)

fahqtut [fully automated high quality translation of unrestricted text] will never happen, boys and girls. Language is too diverse and expressions are often untranslatable. I work in this industry and all the technology I've seen sofar is laughably inadequate. Yes, you can have accurate translations for a restricted range of text, straightforward, present tense stuff sure, it'll work mostly. As soon as you start with the real language that people use your system hits the bricks in a decidedly unflattering manner.
When complex sentences with multiple clauses are used [and this is stuff you're translating, right?] it just cannot keep up. On top of that there's noise in the background, interruptions, people don't enunciate well, use crummy language, use the wrong word altogether, contradict themselves in the same sentence, start sentences they don't complete. This DARPA uber translator thingamajig is going to just come along and happily munch through all that gibberish and deliver crisp and pristine language? Ain't happenin'.

They've been playing with this technology for decades now. Language is our most flexible, most diverse tool. The only way they're ever going to make that work with any measure of reliability is when they figure out how brains work and find a way to build one.

Hey, somebody is going to be working with some exciting technology. I say: let them play.

Language parsing impossible by current technology (2, Insightful)

master_p (608214) | more than 7 years ago | (#16770765)

In order to recognize speech, one needs a context-sensitive parser. In order to make a context-sensitive parser which is fast enough to interpret the text, the computer should have the pattern-matching capacity of a grown up human. The human brain contains 500-1000 trillion synapses! even if one makes the assumption that one synapse equals one bit, in order to understand the context, one would need a computer with a tremendous amount of memory which could be searched in parallel.

Of course if you narrow the problem down to specific terms, then it is doable. But then it would not be 'ultimate' any more.

The English solved this years ago (1)

Colin Smith (2679) | more than 7 years ago | (#16770997)

They just speak a little sslloowweerr and LOUDER! The natives usually catch on.

Politics of scarcity? (1)

seven of five (578993) | more than 7 years ago | (#16771077)

You'd think the FBI would be a prime customer for something like this, but apparently keeping a huge backlog of documents [wnd.com] to translate and a staff that's too small to handle it is more important to the mechanics of their bureaucracy.

The point being, if this tech works, great, but will it be used?

AHA! (1)

paralaxcreations (981218) | more than 7 years ago | (#16771177)

Take that, Wikipedia! [wikipedia.org] It's not just some plot device!

Errr...I mean, soon, it won't be a plot device anymore!

Crap, I mean, eventually it might not be just a plot device...

I mean...oh, fuck it. This is DARPA after all.

Privacy, anyone? (0)

Anonymous Coward | more than 7 years ago | (#16771359)

it could only lead to awesomeness.


Yeah. Computers monitoring all "public" forms of communication - e.g. telephone, e-mail, radio, &c. Key phrase[s] pop up, computer flags it, sends a message to the Thought Police, who come, and bust down your door.


While it may not be practical to have a living human listening in on e.g. every telephone conversation going on at the same time, with government money, one could throw a lot of computing power at this issue. This is DARPA, after all.


I, for one, do not welcome our new, Digital Thought Police, Overlords.

How to Wreck a Nice Beach (2, Insightful)

J.R. Random (801334) | more than 7 years ago | (#16771699)

Just say the title out loud to get some idea of why speech recognition is hard, nevermind translation. Translation has long been regarded as "AI-complete" because to do it well you have to understand what is being said, which involves solving all the other difficult AI problems. The current translation systems are lousy because they don't understand what is being said and most of them don't even attempt to.

So my guess is that this program will be a boondoggle for researchers with little practical result.

You can't impeech his speach :) (1, Funny)

Anonymous Coward | more than 7 years ago | (#16772453)

... speach researchers at Carnegie Mellon University has recently demoed a prototype of a device ...

Real-time translation huh? (1)

gatesvp (957062) | more than 7 years ago | (#16772585)

I'm sure the multi-lingual people out there are laughing at the very concept of "Real-time translation". Unless you're doing something trivial (Italian to Spanish?), this just isn't possible.

Some languages place verbs at the very end of the sentence. Assuming that the computer could understand each of the words, the entire sentence still has to be re-composed in English. For long sentences, the speaker has already moved on.

Other languages, like French, use some crazy sentence structures that effectively do the same thing. A French sentence can be like a string of comma-separated pronouns with the object of the pronoun in the very last bit.

Both of these cases will induce delays and still don't account for some problems of context. In my own personal experience, I've found spots where whole paragraphs really need to be translated if we want to keep the original meaning.

Seriously, I think "Real-time" could be roughly translated :) as within a few seconds.

YOU FAIL IT (-1, Offtopic)

Anonymous Coward | more than 7 years ago | (#16772609)

Will the fansubs be available on torrent? (1)

cylcyl (144755) | more than 7 years ago | (#16772615)

I look forward to them distributing all the translated content on torrent, as a part of the freedom of info act or something, so that we can get English subbed foreign TV programs. I think that the anime shares will be most popular

If it could... (1)

billdar (595311) | more than 7 years ago | (#16772735)

If it could translate what my 11 month old kid is babbling about, it would save both of us a lot of crying... And get me off the sauce.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?