EURASIP Journal on Audio, Speech, and Music Processing

Table 6 Ablation study on the CTC and GRL-based speaker classification network

From: W2VC: WavLM representation based one-shot voice conversion with gradient reversal distillation and CTC supervision

Method	MCD (dB)
W2VC	8.901
w/o CTC	9.287
w/o GRL	9.328
w/o CTC+GRL	9.505

Back to article page