The article describes a series of experiments carried out on an aligned Czech-Lithuanian corpus, in order to extract bilingual dictionaries. Tools that were used for the extraction are described first, and then results for various formats of the input.