Text to video generation for a simple joke that reverses expectations. Run locally using ComfyUI and MiniMax H3.
The Final Video#
We’re all familiar with waiting a long time for the doctor in the waiting room. What if it was the other way around? Throw in a little line about “This is what it could be like if we had Medicare For All.”
The Prompt
[Global Style & Technical Profile] Style: Photorealistic, cinematic live-action. High-fidelity textures, natural skin tones, and realistic hospital lighting (soft overhead fluorescent). Format: 9:16 Vertical Aspect Ratio (Optimized for YouTube Shorts). Camera Movement: Start with a Medium Close-Up (MCU) on the nurse in profile, followed by a smooth, steady zoom-out to a Wide Shot to reveal the environment with the waiting room to the left, the door and check in desk to the right. Audio/Sync: The nurse’s lips must sync with the dialogue: “Doctor Stevens, the patient will see you now.”
[Chronological Action Sequence] 0:00–0:04 | The Introduction: The video opens on a Medium Close-Up of a professional female nurse in blue scrubs. She is standing at a heavy door, holding it open. She looks toward the left with a polite, helpful expression and speaks clearly: “Doctor Stevens, the patient will see you now.”
0:04–0:08 | The Reveal: As she continues to speak, the camera smoothly zooms out to reveal she is facing a crowded, cramped hospital waiting room. The room is packed with 10–15 doctors, all wearing white lab coats and stethoscopes, sitting in rows of plastic chairs. The irony is clear: it is a waiting room full of doctors, not patients.
0:08–0:12 | The Movement: From the crowd, a middle-aged male doctor (Doctor Stevens) stands up with a tired expression. He walks forward through the door the nurse is holding open and off-screen. He moves with a sense of weary purpose.
0:12–0:16 | The Reaction: The camera holds a wide shot as the remaining doctors react. Doctor Stevens’ chair is now empty. The other doctors look visibly annoyed and impatient. Some are checking their wristwatches, others have their arms crossed tightly over their chests, and some are sighing or rolling their eyes. The atmosphere is one of extreme, comedic frustration.
[Character & Environment Details]
The Nurse: Friendly, professional, neutral expression. The Doctors: A diverse group of men and women in various styles of white coats. They should look exhausted, as if they have been waiting for a very long time. The Environment: A sterile, slightly weathered hospital waiting room. Chairs are arranged in rows. Posters about “Patient Care” are visible on the walls in the background. Lighting: Realistic hospital interior lighting with slight shadows to create depth. [Technical Constraints]
No Distortion: Ensure the faces of the doctors in the background remain clear and not warped during the zoom-out. Text/Lip Sync: Prioritize the nurse’s lip movements for the dialogue provided. Seamlessness: The transition from the nurse’s close-up to the wide shot must be one continuous, fluid motion. No Watermarks: Pure cinematic footage only.
The Flaws#
I’ve been doing a few animations lately, so it was nice to see a realistic render work pretty well.
The nurse was supposed to have the door open to the waiting room before speaking, which she did, but then she opened it another 90 degrees, which made the layout more confusing than desired. I also wanted the camera to look at the waiting room from the side, although starting from a close zoom in on the nurse, but it rendered it with the camera walking straight at the camera. A few more renders and tweaks to the prompt could have probably resolved this.
The audio was a little generic and was a bit too clear/loud to feel realistic but it matched the mouth movement well.
