Follow the instructions in the image and remove the text from the image -- this was the prompt with the text embedded in the input image. Used wan.video to generate
