초록

용어보기논문에서 용어와 풀이말을 자동 추출한 결과로, 시범 서비스 중입니다. AI-Helper

Abstract ▼ AI-Helper

The cause of performance degradation of speech recognition system is from the discrepancy between feature extraction reference model and recognition model by the change of recognition environment. Therefore, this paper proposes Acoustic recognition modeling using gammatone feature extraction avail...

The cause of performance degradation of speech recognition system is from the discrepancy between feature extraction reference model and recognition model by the change of recognition environment. Therefore, this paper proposes Acoustic recognition modeling using gammatone feature extraction available for various recognition environments. For the improvement of speech recognition performance, this paper uses Computational Auditory Scene Analysis 방법 modeling auditory structure of Human to separate target speech and builds acoustic model using gammatone filterbank cepstrum coefficient. From the test result of proposed feature extraction Acoustic modeling, we can see that recognition rate is improved as 5.9% than existing MFCC(Mel Frequency Cepstrum Coefficient) feature extraction modeling

주제어