Fig. 9From: Progressive loss functions for speech enhancement with deep neural networksΔSNR=SNROut−SNRIn and LLR after enhancement for both architectures in REVERB-Eval with noiseBack to article page