Audio-Visual and Visual-Only Speech and Speaker Recognition: Issues about Theory, System Design, and Implementation