Abstract. We describe a lexical database consisting of morphologically and phonetically tagged words that occur in the texts primarily used for