Fig. 5From: Heterogeneous separation consistency training for adaptation of unsupervised speech separationSI-SNRi (dB) of DPCCN and Conv-TasNet separation results of the 310 “bad cases” varies with the separation consistency measureBack to article page