In this paper, a Speaker-dependent continuous speech recognition system is described. An unknown utterance is recognized by means of the following procedures: acoustic analysis, phonetic segmentation and identification, word and sentence recognition. In order to adjust some of the system parameters, a speaker adaptation module is able to match known utterances with their acoustical representation. The task to be performed is given as a parameter of the system and is described by its vocabulary and its grammar. Recognition results of continuously spoken sentences extracted from a "pseudo-Logo" language are presented.