Ventures into online machine based translations haven’t met with much success so far – at least not to a degree that can be called phenomenal. While smaller words / phrases / sentences translate fine, the translator engines perform rather pathetically when it comes to large chunks of text.
Translation engines have been around since the early 90’s, when AltaVista came out with its famous Babelfish. In general, the online translators are pretty good and are able to convey the essence of the translated text. However, translator engines fair miserably while operating on non-Western languages specially languages which are based on pictograms / ideograms, like Chinese, Japanese etc. Here’s an interesting explanation by stevej on ZDNet blogs.
Pictographs in the Japanese language do not represent words so much as ideas. A pictograph can have many meanings depending on the context it is used in.
Here’s an interesting exercise. Go to any translation site (you now know of at least two) type in a common English sentence. Translate it to Japanese. Then translate it back to English. Surprised? It doesn’t much resemble your original sentence construct, does it?
While for Western languages the results are much more consistent, they’re far from perfect. They glare at you all the more if you go through a complete translation cycle, i.e. translate from one language to another and then back to the source. Take for instance, the first passage of this article. When converted to Spanish and then back to English, this is how it looks:
The companies in translations automated in line have not satisfied by far success until now – at least not to a degree that can be called phenomenal. Whereas words/phrases/smaller orations translate very well, the motors of the translator are made something pathetically when it comes to the great pieces of the text.
Be it Google or AltaVista or for the matter any other online translation engine of good repute, the results are approximately the same. Attempts to perform literal translations would have borne disastrous consequences. So that’s an option that is out of question. Thus, these engines rely heavily on contextual translations to dig up the correct meaning of a word or phrase. But then again, every language contains commonly used idioms (even in colloquial speech) which cannot be translated keeping the original meaning or connotation intact – even in contextual mode. While performing such a task may come really easy to humans, keep in mind that the complexity and inner workings of a human mind are virtually impossible to emulate using any algorithm. So far we haven’t been able to identify any distinct patterns in the thought process of human brain that can be coded as mathematical algorithms.
Google seems to have finally realised that, following an incessant barrage of whines and grumbles about it’s own translation service over the past few years. Google envisions that in near future …
people will be able to translate documents instantly into the world’s main languages, with machine logic, not expert linguists, leading the way.
In order to make this happen, Google plans to bring some drastic reforms to the algorithms working behind scene. As is expected of a technology pioneer, they’ve devised a new approach to the whole issue – namely, “statistical machine translation” which differs from any of the past efforts in that it forgoes language experts who program grammatical rules and dictionaries into computers.“
The new process involves feeding pre-translated parallel text in various languages into computers and then relying on them to discern patterns for future translations. At the moment the quality offered by this mechanism isn’t perfect either – but there’s a distinct advantage to such a statistical analysis engine. With time, “the more data we feed into the system, the better it gets” said Franz Och, the head of Google’s translation effort at its Mountain View, California headquarters. He further stated that “some people that are in machine translations for a long time and then see our Arabic-English output, then they say, that’s amazing, that’s a breakthrough.“
Franz spends his days feeding hundreds of millions of words from parallel texts such as Arabic and English into the computer, using United Nations and European Union documents as primary sources.
However, according to Miles Osborne, a professor at the University of Edinburgh, who has taken an year-long sabbatical to work on Google’s project – while the renewed efforts on Google’s part are thoroughly praiseworthy, there are inherent limitations to the whole machine-based translation endeavour. At the end of the day, software can never overtake humans in areas of expertise such as this. He compares translations to playing chess. Software at best should be used for understanding rather than polishing documents.
Now it’s just a matter of time and hoping – that Google (or any other company) proves him wrong. After all, it’s going to stand testimony to how well humans understand themselves.
Source of news & quotes of Franz Och from: Google seeks world of instant translations