Machine transliteration refers to the process of automatically converting text from one language into an-other according to the phonetic information.Machine transliteration plays an important role in natural language ap-plications such as multilingual text processing,corpus alignment,and information extraction.This paper investigates the challenges existing in the current machine transliteration tasks,summarizes the main transliteration methods,elaborates the transliteration data sets,and reviews the popular evaluation metrics of transliteration.The existing problems are outlined,and the future research directions of transliteration are discussed.
万方数据
维普
