Self Learning Live Translation System

Overcoming all language ambiguities and achieving machine translation accuracy that is on par with human translation becomes more feasible as more linguistic data is available to developers. This paper outlines such a translation system that is designed to take advantage of massive corpora (prepared by Google and Kaleido K Translation) and is designed on three key fundamentals. Simplicity - to be simple in design to be portable across different languages pairs, Adaptability - to compensate for the evolution of language, and Multiplicity - to determine a final translation from as many candidate translations as possible. Lastly, the translation direction of Korean to English will be used to provide examples and explanations.