Machine learning models can help to solve several real-world problems faster and more efficiently. One among these problems involves synthesizing speech for both animated characters and human speakers based on the movements of their lips.