|This article is an orphan, as no other articles link to it. (February 2009)|
|This article's tone or style may not reflect the encyclopedic tone used on Wikipedia. Specific concerns may be found on the talk page. See Wikipedia's guide to writing better articles for suggestions. (December 2007)|
The technology does away with scheme-based transliteration system that expect users to follow set rules while typing for ex a for अ aa for आ. It uses Artificial Intelligence(Machine learning techniques) in which machine learns different patterns in a language from a training set. This technology has ability to take context (topical/syntactic) into account so words with multiple mappings are also converted correctly.
With dictionary size of over a hundred thousand words, conjuncts, support for proper nouns and foreign language almost everything can be written in Languages of India. One of the most interesting feature of this technology is its predictive transliteration. Predictive transliteration or Auto-Correct feature as we call it, automatically converts to most probable transliteration for the input word. All the user needs to do is type-in the text which he/she wishes to input, and it automatically predicts what the user's typing corresponds to. With so many features it achieves an accuracy of more than 95% for text that includes English words too.
The latest version is only available in Hindi, but older versions can be used for:
| This article uses material from the Wikipedia article Keytrans, that was deleted or is being discussed for deletion, which is released under the Creative Commons Attribution-ShareAlike 3.0 Unported License.