Present systems for computer sign language translation are considered in the article, their advantages and disadvantages are identified. The general case of translation (from sounding Russian into Russian sign language and vice versa) is discussed. A new method for constructing a semantic unit for computer sign language interpretation is proposed. Lexical meanings of words are defined to match the correspondence "word-gesture". For each word certain lexical meaning from a set of alternatives based on the semantic analysis algorithm is confirmed. Semantic analysis algorithms for simple sentences are developed and introduced. A new method for translating Russian text into Russian sign language based on a comparison of syntactic structures is proposed. Special library to determine syntax constructions is developed. Present hardware and software to design the architecture of the prospective system for gesture recognition are considered.
Russian sign language, computer sign language interpretation, semantic analysis, syntax constructions, sign language recognition tools