Improvement of the ASR Robustness using Combinations of Spectral Subtraction and KLT-based Adaptive Comb-filtering

스펙트럴 서브트렉션과 비동기 KLT 잡음 감소 기법의 조합에 의한 음성 인식 성능 개선

In this paper, the combinations of speech enhancement techniques are experimented. Specifically, the spectral subtraction, KLT based comb-filtering, and their combinations are applied to the Aurora2 database. The results show that recognition accuracy is improved when KLT based comb-filtering is applied after spectral subtraction.