Google is developing software for mobile phones capable of translating foreign languages almost instantly. Building on existing technologies in voice recognition and automatic translation, Google hopes to have a basic system ready within a couple of years.

The company has already created an automatic system for translating text on the internet Google Translate. So far Google translate covers 52 languages albeit with various quality.

Google also has a voice recognition system that enables phone users to conduct web searches by speaking commands into their phones rather than typing them in.

So Google is now doing the obvious, trying to combine the two technologies to produce a software capable of understanding a caller’s voice and translating it into a synthetic equivalent in a foreign language.

The software would analyze “packages” of speech, listening to the speaker until it understands the full meaning of words and phrases, before attempting translation.

“We think speech-to-speech translation should be possible and work reasonably well in a few years’ time,” said Franz Och, Google’s head of translation services. “Clearly, for it to work smoothly, you need a combination of high-accuracy machine translation and high-accuracy voice recognition, and that’s what we’re working on.

Although automatic text translators are now reasonably effective, voice recognition has proved more challenging.
“Everyone has a different voice, accent and pitch,” said Och. “But recognition should be effective with mobile phones because by nature they are personal to you. The phone should get a feel for your voice from past voice search queries, for example.”

And that sounds a bit scary if you are interested in privacy not only having Google knowing all your web activities but now also remember your voice…

The translation software is likely to become more accurate the more it is used. And while some translation systems use crude rules based on the grammar of languages, Google is exploiting its vast database of websites and translated documents to improve the accuracy of its system – sounds like the stolen some copies again in China…

“The more data we input, the better the quality,” said Och. There is no shortage of help. “There are a lot of language enthusiasts out there,” he said.

However, some experts believe the hurdles to live translation remain high. David Crystal, honorary professor of linguistics at Bangor University, said: “The problem with speech recognition is the variability in accents. No system at the moment can handle that properly.

“Maybe Google will be able to get there faster than everyone else, but I think it’s unlikely we’ll have a speech device in the next few years that could handle high-speed Glaswegian slang.

Now reading that they will use Google translation does make us believe that it will take years to make something useful.  Although Google Translate works fine in Denmark, German and a few other countries it is almost useless in some Asian countries unless you are a big fan of Scrabble games.

So we translated this article in to Thai and back again in to English for you to see what we think about Google Translate.

Google is developing mobile software can translate foreign languages almost immediately.
Building technology in voice recognition and automatic translation, Google hopes to have infrastructure ready within a few years.
The company has already created an automated system for translation of text in Google Translate Internet. Until now, Google Translate even cover 52 different quality.

Google also has voice recognition system that allows phone users to search the Web by speaking commands in their hands than typing them into.

To make clear that Google is trying to merge two technologies to produce software that can understand and translate the voice call into synthetic equivalent in a foreign language.

Software to analyze "the package" in speaking, listening, speaking, until a full understanding of the meaning of words and phrases before you try to translate.

"We think that to translate spoken words - should be possible and should work well enough in a few years'" Franz Och heads Google's translation services said. "Clearly has to work seamlessly together, you need high-accuracy machine translation and voice recognition, high accuracy and what we are working.

Automatically translate the text but now is the right performance, voice recognition has proved challenging more.
"Everyone has a different voice and accent dark" Och said. But certification will apply to mobile phones because they are by nature personal to you. Call should know the format of your audio sound through the search. the like.

And audio bit frightening if you're interested in privacy, not only is Google to know all your web activity, but now also recognize your voice ...

Translation software should be more accurate with greater use. And while some translations use oil as grammar rules that Google is exploiting the extensive database of sites and document translation to improve the accuracy of the system - sound. such as a copy was stolen some time in China ...

"We put more good quality," Och said. A lack of any help. "There are many enthusiasts of the language out there," he said.

However, some experts believe remains a high hurdle in translation. David Crystal, Honorary Professor of Linguistics at Bangor University, said that problems with speech recognition in noise variance is heavy. System currently can not manage that right.

"Google may be able to travel faster than others, but I think it is unlikely we will have no equipment to speak in the next few years to manage high-speed slang Glaswegian.