How to use jjiang4/acesinger_svs_train_visinger2 with ESPnet:
unknown model type (must be text-to-speech or automatic-speech-recognition)
