It's a minimum mean-squre error short-time spectral amplitude (MMSE-STSA) estimator which is the most poplular statistical model-based technique for speech enhancement.

Time domain	Frequency domain

• [1] Y. Ephraim and D. Malah, "Speech enhancement using a minimum-mean square error short-time spectral amplitude estimator," IEEE Trans Acoustics Speech and Signal Processing, VOL. 32(6):1109-1121, Dec 1984.