Adaptive MMSE speech spectral amplitude estimator under signal presence uncertainty

In this paper, we present an adaptive speech spectral amplitude estimator that minimizes the Mean Square Error (MSE) under signal presence uncertainty. The spectral gain function, that is an adaptive variable-order Minimum MSE (MMSE) estimator, is obtained as a weighted geometric mean of hypothetical gains associated with speech presence and absence. The proposed estimator uses MMSE estimator considering various orders (powers) for the spectrum. The order of estimator and speech presence probability are estimated for each time frame and each frequency component individually. Our evaluations confirm superiority of the proposed method in noise suppression from corrupted speech.