A single channel speech enhancement method based on masking properties and minimum statistics

The paper presents a combined method for the single channel speech enhancement of noisy speech signals at very low signal-to-noise ratios, which is based on masking properties of the human auditory system and power spectral density estimation of non-stationary noise. It allows for an automatic adaptation in time and frequency of the parametric enhancement system, and finds the best tradeoff between a criterion correlated with perception and SNR. This leads to a significant reduction of the unnatural structure of the residual noise. The results show that the enhanced speech is more pleasant to a human listener.