APPARATUS FOR ADAPTING LANGUAGE MODEL SCALE USING SIGNAL-TO-NOISE RATIO

摘要

The present invention relates to a voice recognition system. More particularly, the present invention relates to an apparatus for adapting language model scale to improve voice recognition performance in the voice recognition system. According to the present invention, in case of a voice signal of a low signal-to-noise ratio, a weighted value is applied to the discrimination of a language model. Therefore, recognition performance with regard to noisy environment is improved.