I was reading Cordis and I saw that the EU is funding the “Multilingual on-line translation” tool (a.k.a. MOLTO). It is an ambitious tool but I am afraid that, at least for Greek, we won’t see much of a development. “Why” you would ask?
And I would answer, “Because it states that “The project partners will develop a system with precision and grammar rules in place”. That is the Achilles’ heel for our language. We have a complicated grammar system that, as far as I know, has not been rendered in some viable computational grammar parser.
But then again they say that hope dies last. So here I am hoping that maybe they can ameliorate the Google Translate model (which so far has worked the best out of everything else) and will create something that will prove really useful for Greek. I am including the actual Cordis post along with the source URL in case you want to go to that page and share in Facebook or Twitter the article alone.
EU funds effective translation tool
Source: http://cordis.europa.eu/fetch?CALLER=EN_NEWS&ACTION=D&SESSION=&RCN=31680
[Date: 2010-01-20]
Europeans recognise the importance of communicating in other languages as well as their native tongue, and the availability of effective tools facilitating high-quality translation of texts between multiple languages is pivotal to this. Enter the MOLTO (‘Multilingual on-line translation’) project, which has received over EUR 2.3 million under the ‘Information and communication technologies’ Theme of the Seventh Framework Programme (FP7) to develop such a reliable translation tool.
The MOLTO project partners said the languages will function as separate and varied modules in the tool. The five-strong consortium, which is being coordinated by the University of Gothenburg in Sweden, will develop prototypes that cover most of the EU’s 23 official languages.
In a statement, the university said the MOLTO project seeks to provide the same access to knowledge on the Internet for all EU citizens.
‘It has so far been impossible to produce a translation tool that covers entire languages,’ explained project leader Professor Aarne Ranta from the Department of Computer Science and Engineering at the University of Gothenburg.
A number of online translation tools are currently available to the public. Google Translator, a programme that is used by many people worldwide, for example, gradually improves the quality of translations through machine learning: the system uses feedback to learn from its own mistakes. The disadvantage is that explicit grammatical rules are the exception rather than the rule.
This is where MOLTO comes in. The project partners will develop a system with precision and grammar rules in place. According to the consortium, wide coverage will follow.
‘We wanted to work with a translation technique that is so accurate that people who produce texts can use our translations directly,’ said Professor Ranta. ‘We have now started to move from precision to increased coverage, meaning that we have started to add more languages to the tool and database.’
