Are there any quality voice cloning and speech generation models that already have support in Llama.cpp or, more likely, vLLM-Omni? It would be nice to swap them out like any other inference model and use a common API, rather making a separate container or conda for each model I want to try. MOSS lo
