Abstract
A new version of a large parallel corpus containing translations between a total of 41 languages (including Czech). Compared to version 11, the number of words in foreign texts increased to 1,534 million, including 311 million in the fiction core and 1,223 million in freely available collections.
The total number of words in Czech texts is 200 million, including 111 million in the core and 90 million in the collections. Chinese texts, including POS tags, were added.