Hand gestures have been used as a means of communication among people for a long time, and are interpreted as streams of tokens for language. The sign language is a method of communication for the deaf. This thesis deals with a system which recognizes Korean Sign Language(KSL) and Korean Manual Alphabet(KMA). Real-time recognition of KSL mixed with KMA from a sequence of continuous gestures is very difficult for lack of explicit tokens indicating beginning and ending of signs and for complexity of each gesture. In this thesis, state automata is used for segmenting sequential signs into individual ones. A coding scheme for KSL recognition is proposed using basic elements of KSL and KMA, which consist of 14 hand directions, 23 hand postures and 14 hand orientations. Basic elements are recognized using 3 classifiers implemented by feature analysis using fuzzy rule, and fuzzy min-max neural network(FMMNN). Using a pair of CyberGlove and Polhemus sensor, this system recognizes about 135 Korean signs and 31 KMA's in 15Hz, recognition rate for KMA is 96.7% and KSL is about 95% excluding no recognition case.