How can three people in one picture can speak according to the given text

I have a picture with three people, and I am using the Thunder Client SDK in VSCode. However, when I insert an image featuring three people and assign names and text to each person, the output only displays a video with one person, and all the text is spoken by that one person. I want the generated video to depict all three people speaking when it is their turn in the debate on a topic. Each person presents their point of view. If anyone has code or guidelines, it would be greatly appreciated.