Upload your pictures, write what the narrator reads, and pick the face that reads it.
This face becomes the talking avatar. It is cropped to a square, because the player shows it in a round frame and the video is rendered 1:1.
or
Each page needs one picture and the exact words to read aloud. Pictures are fitted whole into the player's frame, never cropped. Drop several files on the button below to add a run of pages at once.