During source separation, this presumably results in a noisier estimate of the SOI in comparison to mixtures with no additional background noise. We estimated the SNR to be 16.9dB by averaging across multiple samples. The dotted black curve on the left of Figure G.2 is a presumable lower bound on the BER by accounting for the magnitude of the background noise and modeling it as additive white Gaussian noise.

They don't cope well in certain situations and the'transparent' mode lets in unwanted noise

Specifically,we prove that, forasimple data distribution with sparsesignal amidst high-variance noise, a simple convolutional neural network trained using stochastic gradient descent simultaneously learnstothreshold outthenoiseandfindthesignal.

Our method also outperforms baselines in terms of MSE for OFDM mixtures as shown in Figure G.1. The performance at low SIR in the context of CommSignal2 mixtures is not the same.