Possibly the most well-known parallel corpus is the Canadian Hansard Corpus consisting of debates from the Canadian Parliament, that have been published in the country's official languages, English and French. This corpus has been used in research for many years, among others by GaleChurch93 for testing their alignment algorithm as well as Mackl94 in order to find ``fause amis.'' These were then used in the TransCheck System. Their work was based on that part of the corpus that consists of 26 million words recorded from the 1986-1987 debates. Most recently Klavans and Tzoukermann KlavansTzoukermann95 have used the corpus for adding and improving on information in machine-readable bilingual dictionaries.