English-to-Korean Transliteration using Multiple Unbounded Overlapping Phoneme Chunks

We present in this paper the method of English-to-Korean(E-K) transliteration and back-transliteration. In Korean technical documents, many English words are transliterated into Korean words in various forms in diverse ways. As English words and Korean transliterations are usually technical terms and proper nouns, it is hard to nd a transliteration and its variations in a dictionary. Therefore an automatic transliteration system is needed to nd the transliterations of English words without manual intervention. To explain E-K transliteration phenomena, we use phoneme chunks that do not have a length limit. By applying phoneme chunks, we combine di erent length information with easy. The E-K transliteration method has three steps. In the rst, we make a phoneme network that shows all possible transliterations of the given word. In the second step, we apply phoneme chunks, extracted from training data, to calculate the reliability of each possible transliteration. Then we obtain probable transliterations of the given English word.