Mobile menu

Pages in topic:   [1 2] >
The end of our profession?
Thread poster: grzes

grzes  Identity Verified
United States
Local time: 14:16
Polish to English
+ ...
Jul 7, 2006

About a month ago, The Economist published an extremely interesting article. Below are a few excerpts. Take a look and tell me what you think.

"How to build a Babel fish
The Economist, June 8th

Translation software: The science-fiction dream of a machine that understands any language is getting slowly closer.

It is arguably the most useful gadget in the space-farer's toolkit. In “The Hitchhiker's Guide to the Galaxy”, Douglas Adams depicted it as a “small, yellow and leech-like” fish, called a Babel fish, that you stick in your ear. In “Star Trek”, meanwhile, it is known simply as the Universal Language Translator. But whatever you call it, there is no doubting the practical value of a device that is capable of translating any language into another.
Remarkably, however, such devices are now on the verge of becoming a reality, thanks to new “statistical machine translation” software. Unlike previous approaches to machine translation, which relied upon rules identified by linguists which then had to be tediously hand-coded into software, this new method requires absolutely no linguistic knowledge or expert understanding of a language in order to translate it. And last month researchers at Carnegie Mellon University (CMU) in Pittsburgh began work on a machine that they hope will be able to learn a new language simply by getting foreign speakers to talk into it and perhaps, eventually, by watching television.
Within the next few years there will be an explosion in translation technologies, says Alex Waibel, director of the International Centre for Advanced Communication Technology, which is based jointly at the University of Karlsruhe in Germany and at CMU. He predicts there will be real-time automatic dubbing, which will let people watch foreign films or television programmes in their native languages, and search engines that will enable users to trawl through multilingual archives of documents, videos and audio files. And, eventually, there may even be electronic devices that work like Babel fish, whispering translations in your ear as someone speaks to you in a foreign tongue.
This may sound fanciful, but already a system has been developed that can translate speeches or lectures from one language into another, in real time and regardless of the subject matter. The system required no programming of grammatical rules or syntax. Instead it was given a vast number of speeches, and their accurate translations (performed by humans) into a second language, for statistical analysis. One of the reasons it works so well is that these speeches came from the United Nations and the European Parliament, where a broad range of topics are discussed. “The linguistic knowledge is automatically extracted from these huge data resources,” says Dr Waibel.
“Most of the time, the languages that translation researchers deal with in their laboratories are so unfamiliar that they may as well be alien.”
Statistical translation encompasses a range of techniques, but what they all have in common is the use of statistical analysis, rather than rigid rules, to convert text from one language into another. Most systems start with a large bilingual corpus of text. By analysing the frequency with which clusters of words appear in close proximity in the two languages, it is possible to work out which words correspond to each other in the two languages. This approach offers much greater flexibility than rule-based systems, since it translates languages based on how they are actually used, rather than relying on rigid grammatical rules which may not always be observed, and often have exceptions.
Examples abound of the ridiculous results produced by rule-based systems, which are unable to cope in the face of similes, ambiguities or bad grammar. In one example, a sentence written in Arabic meaning “The White House confirmed the existence of a new bin Laden tape” was translated using a standard rule-based translator and became “Alpine white new presence tape registered for coffee confirms Laden.” So it is hardly surprising that researchers in the field have migrated towards statistical translation in the past few years, says Dr Waibel.
Now you're speaking my language
The statistical approach, which starts off without any linguistic knowledge of a language, might seem a strange way of doing things, but it is actually remarkably similar to the way humans attempt to translate languages, says Shou-de Lin, a machine-translation expert who was until recently a researcher at the University of Southern California's Information Sciences Institute (ISI). “It looks at the script and bunches symbols together,” he explains, much as a human mind might try to solve the problem. But in order for this approach to work, the voracious translation systems must be fed with huge numbers of training texts. This prompted Franz Och, Google's machine-translation expert, to boast recently that the search-engine giant would probably have a key role in the future of machine translation, since it has such a huge repository of text....."

cheers
Grzes


Direct link Reply with quote
 
Piotr Wargan  Identity Verified
Poland
Local time: 01:16
English to Polish
+ ...
A true Star Trek Next Generations' vision :-) Jul 7, 2006

Hi,
If they create a TM with all texts ever recorded and compare all language versions, use statistics and then get this device to do its job, then let's beam up to some less technologically advanced world

But.. how will it work in the real world?
A couple of years ago I saw a 'tranlating pen' in a shop in my city. Now I cannot find it any more - why has it disappeared if it was such a great invention - scanning words (or maybe even sentences?) and showing a translation on a small screen.

Don't worry - this monster of a fish may not be so scary very soon!

P.

[Edited at 2006-07-07 19:49]


Direct link Reply with quote
 
Dyran Altenburg  Identity Verified
United States
Local time: 19:16
English to Spanish
+ ...
Resistance is futile Jul 7, 2006

grzes wrote:
About a month ago, The Economist published an extremely interesting article. Below are a few excerpts. Take a look and tell me what you think.
"How to build a Babel fish
The Economist, June 8th
Translation software: The science-fiction dream of a machine that understands any language is getting slowly closer.
uote]

I think a gadget like that would be one of the coolest things ever. Right next to the holodeck, the tricorder, the warp drive, and the replicator. Maybe the stargate too. Still not too sure about that one.

As to your question:


The end of our profession?


I'm not worried at all.

You all have a plan "B", right?

--
Dyran


Direct link Reply with quote
 
Piotr Wargan  Identity Verified
Poland
Local time: 01:16
English to Polish
+ ...
B - yes and this stands for Beam up Jul 7, 2006



[Edited at 2006-07-07 19:48]


Direct link Reply with quote
 

keshab  Identity Verified
Local time: 05:46
Member (2006)
English to Hindi
+ ...
Machine cannot understand Translator's emotion Jul 7, 2006

A dog barks in the same way all over the world. A bird sings the same tune. If any capable translation gadget were translate their voices, it generates same meaning throughout the world. But when it tries on a human being, it must be failed. Why? Because it can translate only material words - it cannot translate the imotion of a human being.

"I know him"- take this simple word. Can any gadget,any machinery,any software explain the mood of the speaker? Whether he is in the mood of sorrow, revenge,laughter or sympathy ? Obviously not. There are many languages in the world which demands special care for each emotion. Which English can do by only three words, extra word for each emotion added in many languages. And who will decide these emotions except mankind? Man is not only logical and intellectual; at the same time emotional too. He can laugh, he can cry, he can overwhelmed with joy which machine cannot !

Therefore friends, don't worry. No gadget can take place of translators until scientists invent a full human being with all emotions in the laboratory!!


Direct link Reply with quote
 
Rosa Maria Duenas Rios  Identity Verified
Local time: 19:16
I will not believe it... Jul 7, 2006

... until I can adequately translate my own texts with such a device. And if it ever turned out to be true, I do have a Plan B ready...

Direct link Reply with quote
 

Timothy Barton
Local time: 02:16
Member (2006)
French to English
+ ...
Yawn.... Jul 7, 2006

Another scary story. MT will never be perfect. It will, of course, improve, but if it does ever get to the point where it's quicker to correct an MT than to translate from scratch (as is already the case for some very closely related languages such as Catalan-Spansh), they'll still need correcting by us, and because of the increased productivity, translation would become more accessible to the general public (ie, more people would be able to afford it) and so there'd be more demand and even more need for translators. For my language combinations though, I still think we are a long way off MT being worth the effort.

Direct link Reply with quote
 

Fan Gao
Australia
Local time: 11:16
Member (2006)
English to Chinese
+ ...
Never...... Jul 8, 2006

but then they do say "never say never" but I'm happy to say it right now:)

Languages are constantly developing. New words are being constantly added. Software can be released but then two weeks later it will be out of date.

No machine can take the place of man or woman in his/her natural environment and keep up to date so regularly with ever adapting and changing languages.

At the end of the day it all comes down to context. I can only speak for English but I have no doubt all other languages have their very same contextual intricacies.

A sentence can be said or written numerous ways but how can a machine ever truly know the context in which it was intended?

Not at all worried in Nanjing:)
Mark


Direct link Reply with quote
 
xxxF Schultze  Identity Verified
United States
Local time: 19:16
Danish to English
+ ...
We've heard about all the marvellous things ... Jul 8, 2006

...that computers will do for us for half a century now, and how they will replace the need for humans to have brains. If you analyze these "improvements" in detail, I'm not so sure they are that great. Take for example a car. 25 years ago if my car broke down on the road, with a background in engineering I had a fair chance of getting it started again. Not so today. It has to be transported to a garage and be connected to a computer. When Bobby Fischer lost to the "Deep Blue" computer, he didn't really lose to a computer. He lost to 4-5 very good chess players who had the help of a huge amount of compute power. I wouldn't say they won. I would call that cheating. Oh, I do not doubt that some day a computer can do a pretty good job of translating the 100% correct source text. The challenge is that you first have to edit the source so it is 100% perfect. Maybe I should be more picky when accepting work, but I find that at least 50% of my time is spent interpreting and editing the source, and less than half of the time is actually spent on translating.

Direct link Reply with quote
 

grzes  Identity Verified
United States
Local time: 14:16
Polish to English
+ ...
TOPIC STARTER
Deep Blue Jul 8, 2006

Flemming Schultze wrote:

When Bobby Fischer lost to the "Deep Blue" computer, he didn't really lose to a computer.


Flemming,
I think it was Kasparov who lost to Deep Blue, not Fischer. I don't think Fischer ever played against Deep Blue, but I might be wrong.
cheers,
Grzes


Direct link Reply with quote
 

jokerman
Germany
Local time: 01:16
English to German
+ ...
Fisher never lost Jul 8, 2006

grzes wrote:

Flemming Schultze wrote:

When Bobby Fischer lost to the "Deep Blue" computer, he didn't really lose to a computer.


Flemming,
I think it was Kasparov who lost to Deep Blue, not Fischer. I don't think Fischer ever played against Deep Blue, but I might be wrong.
cheers,
Grzes


Fisher never played Deep Blue... It was Kasparov who lost... just a small detail, but quite an important one to the chess world

Fischer was the one who played 17...Be6 against D. Byrne in 1956 - a historical move which has been fascinating millions of human beigns for decades. In opposite to that, a machine such as Deep Blue probably wouldn't had been too impressed by that breathtaking move, but that has nothing to do with chess any more.


Direct link Reply with quote
 

Christine Andersen  Identity Verified
Denmark
Local time: 01:16
Member (2003)
Danish to English
+ ...
Never trust 100% matches... Jul 8, 2006

Last week the discussion started from a different angle

http://www.proz.com/topic/50584

- but after my experience of how a text got more and more garbled through using repeat translations, I am convinced that we're going to be needed for a good while yet.

I'm grandmother age, so my personal plan B is retirement

It will take so much time and effort to keep up-dating and re-programming the machines that it will at least be more fun to use human translators for some jobs, and keep the CATs where they belong - as tools, but not running the show.

And what about the bugs and bloopers that creep in - deliberately or not?
I can't bear to think about the malware and anti-malware that would develop ...

When would people ever have time to get away from their computers, eat, sleep, talk to their neighbours, grow flowers and write poetry?

If people don't talk to their children, they won't need translation machines.

And machines can't translate poetry....

My father does maths puzzles in his head for fun. Occasionally he uses the computer to check the results, but it doesn't always understand the problem. If the solution works, then it was correct. If not, he has to redefine something somewhere.

Machines can look at the stars, but they can't ask questions about them. When they run out of facts, they don't get ideas, they stop.

People don't run out of ideas. They keep inventing new words for their new ideas, and other people understand somehow. Machines don't understand. They never will.

Happy translating!


Direct link Reply with quote
 

Andy Watkinson
Spain
Local time: 01:16
Member
Catalan to English
+ ...
Not worried Jul 8, 2006

A couple of thoughts.

Tim says “For my language combinations though, I still think we are a long way off MT being worth the effort.”.

I think it has less to do with language combinations than subject matter.

As we all know, MT is fine when dealing with a limited field, limited vocab., zero requirement to understand the source. e.g. Parts lists, etc…..And in such a case I doubt it matters what the languages involved are.

I think it has more to do with the subject matter.
A sarcastic article in the newspaper, recent medical discoveries, a financial report, a contract (of almost any type)….would you (or anyone) entrust a machine with one of these? Highly doubtful.

As far as chess is concerned, it’s a game with strict rules, unlike language.
Language can reverse the rules in a second. “Impossible is nothing” “I’m bad” (meaning “good”).

Imagine that 5 minutes before a chess match someone decides to change one of the rules. “Pawns must first move 2 squares forward, instead of it being optional” (and no “en passant”)
The human mind would adapt pretty quickly to this and in such a situation even a mediocre player like me could probably beat Deep Blue in 10 minutes.
Its millions of moves, tens of thousands of stored games and strategies would be worthless.

"Not worried in Barcelona"


Direct link Reply with quote
 
Aleksandr Okunev
Local time: 03:16
English to Russian
MT is a translator's tool Jul 8, 2006

Whereas MT developers try to portray it as a
self-sufficient translation system.

I am certain that any MT software will always
require human intervention.

I have ProMT XT which can translate CNC mills
and lathes with an accuracy of 85% fuzzy match.
Took me 2 years and n00k translated words to
achieve this. The field custom dictionary contains
about 5 thousand entries.

More effort will yeild better results, but there will
always be a 'sanity level' - a line, beoynd which
all the fancy machinery just loses its point. Like
hairstyles for instance. Can we make a robot which
will do great hairstyles? Of course we can. Will
anyone will to use it? Hardly ever. There are kinds
of activity where humans will always be irreplaceable
(if there's still some common sense around).

Tool it will remain forever in most fields. And most
of us in this thread will be happily using it by 2010, making more money with less effort.

Stay well,
Aleksandr


Direct link Reply with quote
 
xxxF Schultze  Identity Verified
United States
Local time: 19:16
Danish to English
+ ...
Fischer - Kasparov Jul 8, 2006

Thanks for putting me straight, guys. Not a good sign when you have a better recollection of what happend 50 years ago than that of 5 years ago i guess To tell the truth, chess was never an interest of mine, but the idea of man against machine is. And I will maintain that Kasparov did not lose to a machine. He lost to a team of 10-12 very good chess players who were cheating.

On another tack: Remember the famous phrase (from the Bible): "the spirit indeed is willing, but the flesh is weak".

The anecdote has it that they sent it through an English-Spanish MT and then back through a Spanish- English MT, and it came out like: "The booze was fine, but the meat was rotten".

(And now I am waiting for someone to come back and tell me that it is "on another track"


Direct link Reply with quote
 
Pages in topic:   [1 2] >


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

The end of our profession?

Advanced search


Translation news





memoQ translator pro
Kilgray's memoQ is the world's fastest developing integrated localization & translation environment rendering you more productive and efficient.

With our advanced file filters, unlimited language and advanced file support, memoQ translator pro has been designed for translators and reviewers who work on their own, with other translators or in team-based translation projects.

More info »
WordFinder
The words you want Anywhere, Anytime

WordFinder is the market's fastest and easiest way of finding the right word, term, translation or synonym in one or more dictionaries. In our assortment you can choose among more than 120 dictionaries in 15 languages from leading publishers.

More info »



All of ProZ.com
  • All of ProZ.com
  • Term search
  • Jobs