SPEECH RECOGNITION SYSTEM AND METHOD

摘要

The present invention relates to a speech recognition method which comprises the steps of: extracting a vowel section of the speech; distinguishing vowels by extracting the feature point of lips or tooth area from the video information of user′s face matching with the extracted vowel section of the speech; and recognizing the speech using the information of the distinguished vowels. [Reference numerals] (120) Speech recognition device; (140) Video recognition device; (AA) Voice; (BB) Video