tr_accent reads lines from standard input. Each line contains one word. Every word is looked up in specified dictionaries. All words that when deprived of diacritics match the input word are printed.