PROCEDE DE DETERMINATION DES SIGNAUX DE SEPARATION RESPECTIVEMENT RELATIFS A DES SOURCES SONORES A PARTIR D'UN SIGNAL ISSU DU MELANGE DE CES SIGNAUX

摘要

The method involves determining a signal of estimation of each sound source, and predicting a prediction signal for a current frame based on a separation signal for a previous frame, for each sound source. A separation signal for the current frame is determined by summing the estimated signal and the prediction signal in a weighted manner, where weight factor is found in order to minimize covariance of the separation signal.