APPARATUS FOR RECOGNIZING SPEECH ACCORDING TO DYNAMIC PICTURE AND METHOD THEREOF

PURPOSE: A device and a method of recognizing a voice according to a dynamic screen are provided to reflect screen information of a screen, which is dynamically changed with a screen operation of a user, to voice recognition, thereby increasing the recognition speed and increasing the recognition ra...

Full description

Saved in:
Bibliographic Details
Main Authors HAN, ICK SANG, CHO, JEONG MI
Format Patent
LanguageEnglish
Korean
Published 09.02.2011
Subjects
ACOUSTICS
MUSICAL INSTRUMENTS
PHYSICS
SPEECH ANALYSIS OR SYNTHESIS
SPEECH OR AUDIO CODING OR DECODING
SPEECH OR VOICE PROCESSING
SPEECH RECOGNITION
Online AccessGet full text

Cover

More Information
Summary:PURPOSE: A device and a method of recognizing a voice according to a dynamic screen are provided to reflect screen information of a screen, which is dynamically changed with a screen operation of a user, to voice recognition, thereby increasing the recognition speed and increasing the recognition rate. CONSTITUTION: A voice recognizer includes as following units. A screen information manager(210) manages storing screen information which is changed according to time. A weight controller(220) controls a vocabulary weight and a domain weight according to the change of the screen information. If a voice is inputted, a voice recognition engine(230) obtains the controlled vocabulary weight and the controlled domain weight. The voice recognition engine recognizing a voice using language model, which is dynamically generated.
Bibliography:Application Number: KR20090070833