Pretty close. Not the same kind of animation, but here are some of my experiments with text to video for narrative content:
https://www.youtube.com/watch?v=CgKNTAjQpkk
You'd have to convert to vector, or tweak your model architecture to work with vector format.