Abstract
Most Chinese-English parallel corpora consist of English source texts translated into Chinese. This chapter discusses the need for corpora representative of the under-resourced Chinese into English translation direction. After a brief overview of the current Chinese-English translation scenario and an analysis of existing parallel corpora for this language pair, we discuss problems in mining contemporary Chinese to English translations and issues in Chinese to English parallel text alignment. We then introduce ZHEN, a corpus of circa one-million characters of contemporary simplified Chinese source texts (ZH) from a range of text types aligned with authentic translations into English (EN). Its aim is to contribute to our understanding of Chinese to English translation norms and of features of English translated from Chinese.