SPEECH RECOGNITION SYSTEM AND METHOD BASED ON WORD-LEVEL CANDIDATE GENERATION

摘要

PURPOSE: A voice recognition system and a method thereof based on the candidate generation of a word level are provided to modify a voice recognition result according to each word by changing a corresponding word into a candidate word. CONSTITUTION: A voice recognition result confirmation unit(730) confirms a word string and a candidate word in case candidate word about one or more word is supplied. A word string display unit(740) displays the word string through a display device. The word string display unit classifies one word in the candidate word into the rest word of the word string. [Reference numerals] (710) Voice signal transmitting unit; (720) Voice recognition result receiving unit; (730) Voice recognition result confirmation unit; (740) Word string display unit