CAJ | 학술논문

为了统一地补偿电话语音受加性噪声和卷积通道响应的影响,本文提出了矢量分段多项式近似(VPP)算法.并把此算法成功地应用到稳态噪声和非稳态噪声环境.对于稳态噪声环境,在log谱域采用Batch EM(B-EM)方法;对于非稳态噪声环境,在倒谱域采用递归EM(REM)方法.这两种方法都是基于最小均方误差估计(MMSE)准则的特征补偿.实验结果表明,受背景噪声和电话通道(包括固定电话和GSM)影响的大词汇量连续语音识别应用此算法误识率可以降低约18%.
위료통일지보상전화어음수가성조성화권적통도향응적영향,본문제출료시량분단다항식근사(VPP)산법.병파차산법성공지응용도은태조성화비은태조성배경.대우은태조성배경,재log보역채용Batch EM(B-EM)방법;대우비은태조성배경,재도보역채용체귀EM(REM)방법.저량충방법도시기우최소균방오차고계(MMSE)준칙적특정보상.실험결과표명,수배경조성화전화통도(포괄고정전화화GSM)영향적대사회량련속어음식별응용차산법오식솔가이강저약18%.
A Vector piecewise polynomial (VPP) approximation algorithm is proposed for environment compensation of speech signals degraded by both additive and convolutive noises. By investigating the model of the telephone environment, we propose a piecewise polynomial, namely two linear polynomials and a quadratic polynomial, to approximate the environment function precisely.The VPP is applied either to the stationary noise, or to the non-stationary noise. In the first case,the batch EM is used in log-spectral domain; in the second case the recursive EM with iterative stochastic approximation is developed in cepstral domain. Both approaches are based on the minimum mean squared error (MMSE) sense. Experimental results are presented on the application of this approach in improving the performance of Mandarin large vocabulary continuous speech recognition (LVCSR) due to the background noises and different transmission channels (such as fixed telephone line and GSM). The method can reduce the average character error rate (CER) by about 18%.