I am not 100% sure 'voice morphing' is the correct term. Given a clip or a video with a person speaking, I would like to build a 'voice model' of such a target individual. Then I would like to transform any other voice so that it sounds like the target individual. I am not working in this area but in machine learning. So from a machine learning perspective, I think it should be doable. Can anyone suggest a pipeline or any specific state-of-the-art technique? Thanks.