Yes, we have that technology right now. AI models available today are not as innovative as a human voice actor, but they do a good job of duplicating a basic range of emotions in the actor's cadence.

The question isn't whether we can, but whether the actor's family can sue for using their voice. Right now, performers' rights legislation only protects content made by the actor themselves and does not protect against AI voice duplication. In the coming years, there will undoubtedly be a few legal cases on this subject, but ultimately it may have to be decided on a case-by-case basis.