VOICE RECOGNITION METHOD COMPRISING A TEMPORAL MARKER INSERTION STEP AND CORRESPONDING SYSTEM

摘要

A voice recognition system comprising a decoding stage wherein a word is identified on the basis of voice signal models described with the aid of voice units, each signal model representing a word belonging to a predefined vocabulary. It also comprises step in which the voice signal models are organized into an optimized lexical network associated with rules of syntax, during which each word is identified with a word marker. Temporal information is also inserted into the optimized lexical network in the form of additional generic markers in order to spot relevant moments during the decoding step.