is there any local models that can voice clone, but also supports some sort of expression\emotions on gpu /w 8gb (rtx 4060)?
