Им-----------и-т ------- --пен-ал-?
И__ л_ о___ л___ х______ и п_______
И-а л- о-д- л-с- х-р-и-а и п-н-а-о-
-----------------------------------
Има ли овде лист хартија и пенкало? 0 Im- li -vd-e-l-st -harti-a --pye--alo?I__ l_ o____ l___ k_______ i p________I-a l- o-d-e l-s- k-a-t-ј- i p-e-k-l-?--------------------------------------Ima li ovdye list khartiјa i pyenkalo?
A person who wants to have texts translated has to pay a lot of money.
Professional interpreters or translators are expensive.
Despite this, it is becoming increasingly important to understand other languages.
Computer scientists and computer linguists want to solve this problem.
They have been working for some time now on the development of translation tools.
Today, there are many different programs.
But the quality of machine translations is typically not good.
However, the programmers aren't at fault for that!
Languages are very complex structures.
Computers, on the other hand, are based on simple mathematical principles.
Therefore, they can't always process languages correctly.
A translation program must learn a language completely.
In order for that to happen, experts have to teach it thousands of words and rules.
That is practically impossible.
It's easier to have a computer crunch numbers.
It's good at that!
A computer can calculate which combinations are common.
It recognizes, for example, which words are often next to each other.
For that, it has to be given texts in various languages.
This way it learns what is typical for certain languages.
This statistical method will improve automatic translations.
However, computers cannot replace humans.
No machine can imitate a human brain when it comes to language.
So translators and interpreters will have work for a long time to come!
In the future, simple texts could certainly be translated by computers.
Songs, poetry and literature, on the other hand, need a living element.
They thrive on the human feeling for language.
And it's good that way…