Fig. 6From: Motor data-regularized nonnegative matrix factorization for ego-noise suppressionEstimated ego-noise activations by a NMF and b motor data-regularized NMF for time frames ℓ and j, where ℓ and j were chosen such that Wℓj is large (here Wℓj≈0.9). The considered mixture of ego-noise (Sequence I) and speech has SNR = 3 dB, dictionary sizes K=20 and KS=20. For motor data-regularized NMF, λT=0.9, λE=19 were usedBack to article page