A worrying development? Thread poster: Ellie B
|
|---|
Ellie B 
United Kingdom
Local time: 21:14
French to English
| ATIL KAYHAN
Türkiye
Local time: 23:14
Member (2007)
Turkish to English
+ ...
| It is not worrying at all... | Nov 9, 2012 |
The so called machine translation has been an ongoing issue for ages. Microsoft's English into Mandarin translation is not a new development if you ask me. Translation contains elements of both art and science. A computer can perfectly take care of the science part. What about the art component? This is something that still needs the human touch. A computer cannot imitate art. For example, a computer cannot write a poem (as far as I know). For that reason, we do not need to worry about c... See more The so called machine translation has been an ongoing issue for ages. Microsoft's English into Mandarin translation is not a new development if you ask me. Translation contains elements of both art and science. A computer can perfectly take care of the science part. What about the art component? This is something that still needs the human touch. A computer cannot imitate art. For example, a computer cannot write a poem (as far as I know). For that reason, we do not need to worry about computers replacing humans in translation. They might serve humans as cheap and imperfect translators but that is all. ▲ Collapse
| | | | Phil Hand
China
Local time: 04:14
Chinese to English
| Not as good as it looks | Nov 9, 2012 |
The bits where you can hear the Chinese aren't terrible. But they're not good, either. It does demonstrate the gradual sharpening of these technologies - computer generated voices are much better than they used to be, and the ability to translate basic words correctly is improving sharply. But the meaning... there's a long way to go.
| | | | Neil Coffey
United Kingdom
Local time: 21:14
French to English
+ ...
| An interesting party trick at this stage | Nov 9, 2012 |
This is a fascinating area of research and one day in the future, it may well be that much of the humdrum interpreting task itself is actually executed by machines, while the task of the human linguists shifts to working behind the scenes on improving the system.
And I actually think that Atil's comment about "art" vs "science" is really a reflection of *current* technology, not that in principle machines couldn't capture much of the "art".
*But*... be that as it may, w... See more This is a fascinating area of research and one day in the future, it may well be that much of the humdrum interpreting task itself is actually executed by machines, while the task of the human linguists shifts to working behind the scenes on improving the system.
And I actually think that Atil's comment about "art" vs "science" is really a reflection of *current* technology, not that in principle machines couldn't capture much of the "art".
*But*... be that as it may, we clearly have a way to go. At the moment, this type of demonstration is really at the level of an amusing party trick.
What is interesting, however, is that the industry is clearly drooling over a system with characteristics which would be thrown out as unfit for purpose if a human interpreter had those characteristics. Imagine telling a client that you will be performing interpretation where "the error rate is one word in 7 or 8": do you think they would be nearly so enthusiastic? Imagine assigning marks to an interpreter who misheard "than we previously were able to do" for "and we previously if we're able": would you still be clapping at their every word as this enthusiastic crowd is doing to the computer voice?
If this were a human interpreter, you wouldn't think that their performance was "business ready". Yet as you can see, the industry is incredibly hasty in attributing such a label to similarly shoddy results so long as those results are achieved by a computer. ▲ Collapse
| | |
|
|
|
Neil Coffey
United Kingdom
Local time: 21:14
French to English
+ ...
| So, to answer the original question... | Nov 9, 2012 |
I'm not "worried" by the development per se: au contraire, it is going to open up all sorts of new technology and opportunities in the long run. But in the short run at least, I worry that we are going to see a lot of linguistically ignorant technologists enthuse about this technology and thereby reinforce one another's ignorance.
What is clear is that at the moment we have a contradiction: we are increasingly making use of speech technologies, yet linguistics-- as opposed to, say, ... See more I'm not "worried" by the development per se: au contraire, it is going to open up all sorts of new technology and opportunities in the long run. But in the short run at least, I worry that we are going to see a lot of linguistically ignorant technologists enthuse about this technology and thereby reinforce one another's ignorance.
What is clear is that at the moment we have a contradiction: we are increasingly making use of speech technologies, yet linguistics-- as opposed to, say, physics or biology-- is still not seen as a "mainstream" science. It would really help if we started trying to make sure that the public at large has enough basic knowledge about language as a whole to be able to assess how suitable these technologies really are for their purpose. ▲ Collapse
| | | | | Secretaries worrying about computers | Nov 9, 2012 |
When computers landed in companies, secretaries were terrified of losing their typewriters. Have secretaries disappeared? Not at all. Will translators disappear with automatic translation/interpreting? Not at all. It will in fact promote a higher degree of globalisation... and a bigger need for professional human translation.
| | | | | Why do we still worry? | Nov 9, 2012 |
I wonder why we keep worrying with every technological development in our industry: typewriters, computers, CAT tools, automatic translation, automatic OCR+translation, automatic interpreting... So far the industry has not shrunk at all even with all these advancements, but instead has grown consistently and seems to have a bright future ahead.
I simply refuse to keep worrying!! Let's get to work people and exploit technology in our favour!!
| | | | | I think it's good news | Nov 9, 2012 |
Every day I get a message via my web site from some countryfellow of mine who:
- has a book they must read for their master's course;
- has a book a cousin lent them with glowing recommendations; or
- bought some hi-tech thingamajig on their last trip to Miami with the instruction manual
... which are all in English, and they only speak Portuguese.
I ex... See more Every day I get a message via my web site from some countryfellow of mine who:
- has a book they must read for their master's course;
- has a book a cousin lent them with glowing recommendations; or
- bought some hi-tech thingamajig on their last trip to Miami with the instruction manual
... which are all in English, and they only speak Portuguese.
I explain them that a book publisher or a gizmo importer will be able to spread the translation cost over a few thousand books/units they'll sell, however professional translation is too expensive for an individual user.
Actually, an ESL course will often cost them less than one such translation, however it would take much, much longer.
I give them my price, anyway, and advise them that they might find amateurs who would do it for half that amount, or even less. However their quality is usually so bad, that they'll be better served by the free online Google Translate (I provide the URL), in spite of the tons of nonsense they'll possibly get.
So I think that good machine translation will cover an actual need these people have, and also force amateur-level translators to either develop to a fully professional quality level, or seek some other endeavor to make a living. This should, to a reasonable extent, cleanse the translation market.
Quite honestly, I don't envision machine translation doing the work I do. While some automatic transmissions can shift gears as smoothly as I do with a stickshift, translation is much, much more complex than that, and will remain so. I'm not worried at all.
[Edited at 2012-11-09 21:11 GMT] ▲ Collapse
| | |
|
|
|
Phil Hand
China
Local time: 04:14
Chinese to English
| I'm not convinced that linguistics has contributed anything | Nov 9, 2012 |
Neil Coffey wrote:
What is clear is that at the moment we have a contradiction: we are increasingly making use of speech technologies, yet linguistics-- as opposed to, say, physics or biology-- is still not seen as a "mainstream" science. It would really help if we started trying to make sure that the public at large has enough basic knowledge about language as a whole to be able to assess how suitable these technologies really are for their purpose.
When I was at university I did linguistics, and I went out with a girl who was doing computer linguistics, and the two fields were almost completely unrelated. In this video, pretty much the only technical term mentioned is "hidden Markov model". What's really fascinating about the current generation of language software tools is that they *don't* incorporate any linguistic insights. They do it just by number crunching. If they were any good, I'd say this constituted an argument against a Chomskyan UG or language faculty - but as they're not, I don't think it means much, except perhaps that descriptive linguistics hasn't actually got very far.
So I certainly wish the public knew more about language. And I love the discipline of linguistics considerably more than the next man. But I think we should be very cautious when talking about its achievements. There's been an admirable breadth of research done, with descriptions of many languages; but in terms of depth, I don't think linguistics has got anywhere at all yet.
| | | | Ty Kendall
United Kingdom
Local time: 21:14
Hebrew to English
| Computational Linguistics / Linguistics | Nov 10, 2012 |
Phil Hand wrote:
When I was at university I did linguistics, and I went out with a girl who was doing computer linguistics, and the two fields were almost completely unrelated.
You've just reminded me of my time at university. The Computational Linguistics Dept tried to indoctrinate me into their circle for a time,...I did some brief 'researchy' work for them before coming to the realization that it was far too "computational" and not enough "linguistics" for me to be interested in.
(Ultimately, a friend of mine ended up becoming their PhD slave,...definitely dodged the bullet there!)....and I even avoided the upcoming elective "Computational Linguistics" module altogether and did something else instead.
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » A worrying development? | Trados Studio 2022 Freelance | The leading translation software used by over 270,000 translators.
Designed with your feedback in mind, Trados Studio 2022 delivers an unrivalled, powerful desktop
and cloud solution, empowering you to work in the most efficient and cost-effective way.
More info » |
| | Protemos translation business management system | Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!
The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news
