Coffee and Serendipity is an AI-generated dialogue video, with characters speaking on screen, made with TubeTube in the Photorealistic film visual style, 0:42 long across 7 scenes. Scene images were generated with GPT Image 2.5 · Low and animated with Veo 3.1 Fast · 720p. The full recipe below is remixable: reuse the same style and settings with your own words.
Scenes 7 frames
The exact frames generated for this video - remix the recipe to make your own with the same look.
The script as spoken in the video
[A sunlit, modern café.] Narrator: An ordinary Tuesday morning, quiet and unassuming.
[Maya sits by the window.] Narrator: Maya sits, a pen slipping from her open notebook.
[The pen rolls onto the tiled floor.]
[Dev steps over, picking it up.] Dev: Excuse me, I think you dropped this.
[Maya looks up at him.] Maya: Oh, thank you. I was miles away just now.
[Dev smiles warmly.] Dev: It happens before the first coffee. Have a good morning.
[Maya smiles back.] Maya: You too.
The prompts copy anything
Image style
Cinematic PHOTOREALISTIC style, like a real live-action film: ultra-detailed 8K photo quality, volumetric cinema lighting, shallow depth of field, lifelike characters with natural skin and fabric textures, epic color grading.
Scene prompt example 1
A quiet, sunlit, modern café on an ordinary morning. Sunlight streams through large windows, illuminating empty modern wooden tables and chairs. Peaceful atmosphere, no people, photorealistic, cinematic lighting.
Scene prompt example 2
A beautiful South Asian woman in her late 30s with warm olive skin, long thick dark brown wavy hair, wearing a stylish muted terracotta silk blouse and delicate gold chain necklace, sits alone by a window in a sunlit modern café. An open notebook rests on the table, and a pen is captured mid-air, falling from the notebook. Her lips are closed, she is not speaking. Cinematic, quiet elegance.
Scene prompt example 3
Close-up macro shot of a pen rolling across a clean, modern tiled floor inside a sunlit café. Warm morning sunlight casts reflections on the tiles. No people are visible. Highly detailed, photorealistic.
Scene prompts are written automatically from your lyrics - these are examples from this video.
