AN EM-BASED APPROACH FOR PARAMETER ENHANCEMENT WITH AN APPLICATION TOSPEECH SIGNALS

This paper considers the estimation of signal parameters and their enh ancement using an approach based on the estimation-maximation (EM) alg orithm, when only noisy observation data are available. The algorithm is derived with an application to speech signals. The distribution of the excitation source for the speech signal is assumed as a mixture of two Gaussian probability distribution functions with differing varian ces. This mixture assumption is experimentally valid in enhancing nois e-corrupted speech, We recursively estimate the signal parameters and analyze the characteristics of its excitation source in a sequential m anner. In the maximum likelihood estimation scheme we utilize the EM a lgorithm, and employ a detection and an estimation step for the parame ters. For their enhancement we use a Kalman filter for the parameters obtained from the estimation procedure, Simulation results using synth etic and real speech data confirm the improved performance of our algo rithm in noisy situations, with an increase of about 3 dB in terms of output SNR compared to conventional Gaussian assumption. The proposed algorithm also may be noteworthy in that it needs no voiced/unvoiced d ecision logic, thanks to the use of the residual approach in the speec h signal model.