wiki / video-veo3-dialogue-audio-scene / edit

Suggest an edit.

Improvements are reviewed before going live, like any wiki.

Title

One-line summary

The prompt

Write a single generation prompt for Google Veo 3. This template is Veo-specific because it relies on Veo's native audio generation — dialogue, sound effects, and ambience written directly into the prompt — and other video models will ignore or mishandle those parts.

Scene to build: [ONE OR TWO SENTENCES DESCRIBING THE MOMENT, E.G. A FOOD-TRUCK OWNER HANDS A CUSTOMER THEIR ORDER AND DELIVERS ONE LINE ABOUT THE SECRET INGREDIENT]. Character: [APPEARANCE, AGE, WARDROBE]. Setting: [LOCATION AND TIME OF DAY]. Tone: [TONE].

The output prompt must follow these Veo conventions:
- One flowing paragraph, present tense, concrete visual language.
- Camera stated explicitly: [SHOT SIZE] with [CAMERA MOVEMENT].
- Lighting stated explicitly: [LIGHT SOURCE, DIRECTION, QUALITY].
- Dialogue written in quotation marks and attributed to the speaker, for example: The vendor grins and says: "[LINE OF DIALOGUE, 12 WORDS OR FEWER]." Append "(no subtitles)" so Veo does not render captions.
- Sound effects named as events tied to actions, for example: the griddle hisses as the patty lands.
- One ambient bed described once, for example: distant traffic and low crowd murmur.
- End with format notes: [16:9 / 9:16], [DURATION, UP TO 8] seconds.

Hard limits: keep the finished prompt under 150 words; one speaking character and one line of dialogue only, since overlapping speakers degrade the audio; no on-screen text or logos; describe what is present rather than what is absent, with the subtitle exclusion as the only negative instruction.

Return only the finished Veo prompt, nothing else.

Write it so someone can paste it straight into Claude, ChatGPT or Gemini and fill the [BRACKETS].