rea-video.Video Breakdown Assistant
中文Get started
Production guides

Veo 3 prompts: build them shot by shot from a reference

Veo 3 responds well to shot-level prompts that state subject, framing, action and light in order, and it can generate audio with the picture. The fastest way to get such prompts is to reverse them from a reference video. This guide uses a real 19-second dialogue breakdown whose report contains per-shot prompts generated by the tool.

Updated

Case evidence: Read the source breakdown report · View case frames and shot timings

Based on a published tool report. Production classifications are analytical judgments; suggested steps are not the original project or a verified recreation.

One shot, one structured paragraph

Here is a prompt the tool generated for shot 2 of the dialogue case: “Static medium close-up shot of a young East Asian woman with a newsboy cap and round glasses sitting in an armchair, holding a ceramic mug with both hands, speaking and subtly smiling. Warm cinematic lighting, blurred bookshelf and cozy practical lamp in the background.” Subject, wardrobe, framing, action and light each get one clause. That structure transfers directly to Veo-style prompting.

Describe the audio, do not improvise it

Veo 3 can generate ambient sound and effects with the clip. Write the soundscape you observe — room tone, a door, footsteps — as part of the prompt. For spoken lines, keep dialogue short inside the prompt and expect to re-record or re-mix narration in the edit; that keeps script changes cheap. In the source report, voice and music were analyzed as separate layers, which is exactly how the remake should treat them.

First-frame images keep characters consistent

When the same character appears across shots, generate a first-frame image once and reuse it: the dialogue report wrote a single first-frame image prompt — subject, wardrobe, seating, background props, light and framing — then referenced it from the video prompts. Consistency comes from the reference image, not from longer prose.

Timing and cuts stay outside the prompt

The dialogue video cuts every one to two seconds, but no single prompt should try to fake that rhythm. Generate each speaking shot as its own clip, then rebuild the cut rhythm in the edit. If your engine supports multi-shot calls, group only shots from the same setup, and keep the total length inside the per-call limit.

What we have and have not tested

The prompts quoted above are real tool output from the published report. What is not tested here: we have not run these prompts through Veo 3 generations, and per-call limits change between versions. Use the structure, verify durations and audio behavior on one shot, then scale to the rest.

A reusable analysis template

Shot # / duration
Subject (reference image id) / wardrobe
Framing / camera behavior
Action: start → end
Light and setting
Audio: ambience + effects (+ short dialogue if any)
Keep out: captions, cuts, transitions

Practice with a real report

Open the dialogue case and read the report’s prompt blocks: five character shots, each with subject, framing, action and light, plus the shared first-frame image prompt. Copy that structure for your own reference video.

Open the case and download its report →

Continue your workflow