Browse > Article

Frame Reliability Weighting for Robust Speech Recognition  

조훈영 (한국과학기술원 전산학과)
김락용 (한국과학기술원 전산학과)
오영환 (한국과학기술원 전산학과)
Abstract
This paper proposes a frame reliability weighting method to compensate for a time-selective noise that occurs at random positions of speech signal contaminating certain parts of the speech signal. Speech frames have different degrees of reliability and the reliability is proportional to SNR (signal-to noise ratio). While it is feasible to estimate frame Sl? by using the noise information from non-speech interval under a stationary noisy situation, it is difficult to obtain noise spectrum for a time-selective noise. Therefore, we used statistical models of clean speech for the estimation of the frame reliability. The proposed MFR (model-based frame reliability) approximates frame SNR values using filterbank energy vectors that are obtained by the inverse transformation of input MFCC (mal-frequency cepstral coefficient) vectors and mean vectors of a reference model. Experiments on various burnt noises revealed that the proposed method could represent the frame reliability effectively. We could improve the recognition performance by using MFR values as weighting factors at the likelihood calculation step.
Keywords
Frame reliability; Reliability weighting; Frame SNR; Burst noise; Robust speech recognition;
Citations & Related Records
Times Cited By KSCI : 1  (Citation Analysis)
연도 인용수 순위
1 Degraded word Recognition based on Segmental Signal-to-Noise Ratio Weighting /
[ H. Kobatake;Y. Matsunoo ] / Proc. IEEE Int. Conf. Acoustic Speech Signal Processing
2 Speech recognition based on variable informations rate model /
[ I.J. Choi;C.K. Un;N.S. Kim ] / Electronics Letters   DOI   ScienceOn
3 음성데이터베이스의 현황 및 과제 /
[ 이용주 ] / 제13회 음성통신 및 신호처리 워크샵
4 손실 데이터 이론을 이용한 강인한 음성인식 /
[ 김락용;조훈영;오영환 ] / 한국음향학회지   과학기술학회마을
5 Speech Recognition in Noise enviroments;A Survey /
[ Y. Gong ] / Speech Communication   DOI   ScienceOn
6 Weighted Matching Algorithms and Reliabilty in Noise Cancelling by Spectral Substraction /
[ N.B. Yoma;F. Mclnnes;M. Jack ] / Proc. IEEE Int. Conf. Acoustic Speech Signal Processing
7 Robust automatic speech recognition with missing and unreliable acoustic data /
[ M. Cooke;P. Green;L. Josifovsk;A. Vizinho ] / Speech Communication
8 Speech Recognition in Impulsive Noise /
[ S.V. Vaseghi;B.P. Milner ] / Proc. IEEE Int. Conf. Acoustic Speech Signal Processing
9 The use of variable frame rate analysis in speech recognition /
[ K.M. Ponting;S.M. Peeling ] / Computational Speech and Language   DOI