Figure 3From: An iterative model-based approach to cochannel speech separation Input SNR estimation error (in dB) with respect to number of iterations used in iterative estimation. Back to article page