METHOD AND APPARATUS FOR SPEECH RECOGNITION

摘要

Disclosed are a voice recognition method and a voice recognition apparatus. The disclosed voice recognition method generates a word sequence on the basis of a phoneme sequence generated from a voice signal, and generates a syllable sequence corresponding to one of multiple words included in the word sequence on the basis of the phoneme sequence, and can determine a text, a recognition result of the voice signal, on the basis of the word sequence and the syllable sequence. The voice recognition method of the present invention can increase accuracy of voice recognition effectively.