How to use espnet/amuse_encodec_16k with ESPnet:
unknown model type (must be text-to-speech or automatic-speech-recognition)
