Google’s technology is evolving quite rapidly and search is not the only place where the company is making big strides. The pace of technological innovations has placed Google ahead of every other major company in a highly competitive market. Not to say that the niche players are sleeping, but with its demonic size and sky touching stock prices, buying out competition is not a problem for Google. In web advertising, Google is already the undisputed leader, the doubleclick acquisition even irked many a competitors and industry watchers alike. Not so long ago, Google announced a bunch of new features at its Google Searchology event, then at the Google I/O it was Google Wave and now it’s Google Translator Toolkit.

Google, on 9th June, 2009 announced the launch of the innovative Google Translator Toolkit which will add human touch to the machine translation. We all are aware of this fact that translation when done online on any website using any translator software or interface (even its own Google’s Translator) is flawed. The words do get translated but their context remains the same (of their original language) which makes the translation full of flaws as their meaning changes or the translation gets absurd.

To remove this imperfection and make the translation meaningful, the new translator toolkit is an option to reckon with. It is a powerful yet easy to use editor that adds a human touch to the translation. Michael Galvez and Sanjay Bhansali of Google Translator Toolkit team explained the functioning of translator toolkit:

“For example, if an Arabic-speaking reader wants to translate a Wikipedia™ article into Arabic, she loads the article into Translator Toolkit, corrects the automatic translation, and clicks publish. By using Translator Toolkit’s bag of tools — translation search, bilingual dictionaries, and ratings, she translates and publishes the article faster and better into Arabic. The Translator Toolkit is integrated with Wikipedia, making it easy to publish translated articles. Best of all, our automatic translation system “learns” from her corrections, creating a virtuous cycle that can help translate content into 47 languages, or over 98% of the world’s Internet population.”

At present, the new translator toolkit is integrated with Wikipedia and Google Knol. The user can use his/her Google or Gmail ID to login to the service and execute the translation process. The interface of Google Translator Toolkit is quite clean with only a few options making it easy for novice users to understand the service.

The translation process is quite simple. The user has to simply upload the file to be translated by clicking the “Upload” button. At present, the system supports following types of local files:

•    HTML (.html)
•    Microsoft Word (.doc)
•    OpenDocument Text (.odt)
•    Plain Text (.txt)
•    Rich Text (.rtf)

The unique feature of the translator toolkit is its Translation Memories (TM). Well, this term isn’t the rocket science. Translation Memories simply mean the database of human translations. As you translate the sentences, the toolkit automatically searches for previous translations similar to the new ones. If the match exists, they are shown to the user which can be reused by them thereby saving time and also avoiding unnecessary duplication. By default, all the translations are set to “Global, shared TM” i.e. they are placed in the publicly searchable translation memory.

In cases where the user might not want to make their translations publicly available, an option of private translation memory is present. This is how it can be executed:

1.    When creating a translation, click Sharing.

2.    Type the name of the translation memory where you want to save the translations. If you have no translation memories, you can create one by clicking on the “create a TM” link.

The above specified custom translation memories are not supported for Wikipedia and Knol. The reason behind it is quite obvious. The content on Wikipedia and Knol are meant to be shared with every user and therefore all the translations made in these can only be saved to “Global, shared TM”.

Some users may want a glossary to be associated with the translation process from which a few terminologies can be extracted. In such cases, the user can associate a glossary with the translation process. Here the user can either create a new glossary or associate the translation with the existing one.

The Google Translator Toolkit follows certain protocols to search for and rank the translation memories. All else being equal, primary translation memories search results trump collaborative translation memories search results. Similarly, collaborative TM search results trump general TM results.

•    If a translation has an associated, primary TM, the toolkit searches that TM first for previous translations of your sentences.

•    If the toolkit doesn’t have enough relevant results from the primary TM, it searches all the TMs to which you have access through collaboration.

•    If it still doesn’t have enough relevant results from the primary and collaborative TMs, it searches the general, shared TM.

The new translator toolkit is definitely going to revolutionize the translation process and will further the cause of Google’s mission which is, to organize the world’s information and make it universally accessible and useful. Translation in our view is essential to accomplish this mission. Moreover, as the Global Translation Memories (TM) grows, it might even be possible for the users to get near-to-human quality translation for their documents, blogs, articles, web-pages or emails or just about anything.