Syntax-directed recognition of spoken words in real-time

A real-time recognizer of isolated words spoken in the italian language is presented. Each distinct level of recognition (segmentation, phonemic classification and lexical recognition) is controlled by the rules of appropriate grammars whose symbols are fuzzy linguistic variables. The recognition strategy depends on the lexical redundancy of the protocol and is based on a classification of speech units into broad phonetic classes, eventually followed by a classification into more detailed classes if some ambiguities still remain. The recognition steps are shown in detail through some examples.