How to Build AI Video Prompts in NadouPro

To build AI video prompts in NadouPro, write them like a production brief: define the subject, visible action, scene, cinematic language, lighting, mood, reference inputs, and any audio or dialogue needs. In NadouPro, this structure can be used for the Video node, expanded with Prompt Optimization, strengthened with @ references from text, images, audio, or video, and organized on Canvas or Workbench to create a clearer path from script and storyboard to final edit.

A simple prompt structure for NadouPro video generation

A good AI video prompt should tell the model what the audience should see and hear, not just what the creator has imagined. The most reliable structure is to break the prompt into production decisions instead of writing one very long abstract sentence.

Prompt section	What to specify	Example phrasing patterns
Subject	The main character, product, prop, or environment	A young architect in a dark coat, a glass perfume bottle, an empty subway platform
Action	Visible movement or behavior	Walks toward the camera, slowly rotates the product, opens the door and pauses
Scene	Location, time, and physical details	Inside a rain-covered city alley at night, in a warm minimalist studio
Shot	Shot size, angle, and movement	Medium Shot, Low Angle, slow Dolly In, Static Shot, Orbit Left
Lighting	Key Light, Rim Light, softness, direction, contrast	Soft Key Light from the left, subtle Rim Light, dark background shadows
Mood	Emotional tone and genre direction	Quiet suspense, the calm feel of a refined commercial, nostalgic warmth
References	Images, video, audio, or text used as inputs	Use the clothing from @image1 and the movement rhythm from @video1
Audio or dialogue	Sound, line delivery, and voice intent	Use Voice Dubbing for this line, then apply Lip Sync to match the speaker

A reusable NadouPro prompt template

When you want a clean, controllable starting point, use this template:

Subject: [who or what is the focus]. Action: [visible movement]. Scene: [location, time, important physical details]. Shot: [shot size, angle, camera movement]. Lighting: [light source, texture, contrast, direction]. Mood: [emotional tone]. References: Use [@image, @video, @audio, or @text] for [specific purpose]. Output intent: [single shot, transition, product segment, dialogue beat, or storyboard segment].

Make every section useful. A short prompt with clear decisions is usually easier to control than a long prompt filled with repeated adjectives.

How to write cinematic language into a prompt

Cinematic language should describe how the camera directs the audience. NadouPro supports standard film terms in prompts, including shot sizes such as Close-Up, Medium Shot, Full Shot, Wide Shot, Over-the-Shoulder Shot, Point-of-View Shot, and Environment Shot; camera angles such as Eye Level, High Angle, Low Angle, Dutch Angle, Bird's-Eye, and Worm's-Eye; and movements such as Dolly In/Out, Pan Left/Right, Tilt Up/Down, Crane Up/Down, Orbit Left/Right, Tracking Follow/Lead, Dolly Zoom, and Static Shot.

Shot prompt pattern

For clearer execution, write shot instructions in this order:

1. Starting frame: Where the shot begins and what fills the frame.
2. Movement: One main camera movement, such as a slow Dolly In or Tracking Follow.
3. Ending frame: What the shot finally reveals or emphasizes.
4. Reason: The narrative purpose, such as revealing product details or highlighting a character reaction.

Example: Begin with a Medium Shot of a chef placing a ceramic bowl on a wooden countertop. The camera slowly Dolly In toward the steam rising from the bowl, finally stopping on a Close-Up that shows texture and garnish. Warm soft Key Light from camera left, gentle Rim Light along the bowl edge, creating the calm mood of a high-end food commercial.

How to use references in NadouPro prompts

NadouPro's node system supports @ references, so a prompt can combine text, images, audio, and video as inputs. This is useful when text alone is not enough to preserve character, product, scene, movement, or sound direction.

• Use image references when visual identity is important, such as product shape, clothing, faces, props, or scene style.
• Use video references when action, gesture, or rhythm is important.
• Use audio when the Video node or an audio workflow needs sound input, such as using a dubbing track for Lip Sync.
• Use text references when a script, shot list, or brand brief should guide generation.

To maintain character consistency, prepare stable reference materials first. NadouPro supports Image Fusion, Multi-reference video generation using 1–7 related images, Character Body 3-View, Character Face 3-View, and Character Expression 9-Grid. These materials can help the same subject remain recognizable across different shots, although the final result still needs review.

Where NadouPro tools fit into the prompt workflow

NadouPro is not just a prompt input box. It provides multiple workflow tools that help turn prompts into a structured production process.

• Video node: Use Text to Video, Image to Video, First/Last Frame, Multi-reference, Frame Interpolation / Ultra HD, and Erase in suitable scenarios.
• Canvas: Connect nodes and use @ references to combine materials across text, images, audio, and video.
• Workbench: Organize a more production-oriented process, from script to storyboard to production.
• Storyboard node: Convert script text into storyboard cards when you need shot planning before generation.
• 3D Director: Use 3D scene snapshots and reference exports to plan character blocking and camera arrangements, especially when a sequence needs clearer staging.
• Prompt Optimization: Expand or refine rough ideas into prompts that are closer to production-ready.

How to handle dialogue and performance

If the video needs lines, separate the visual prompt from the audio workflow. NadouPro includes Voice convert and Text to speech. A practical workflow is to generate or prepare the voice first, then use Lip Sync to align the character's mouth movement with the dubbing audio.

When to use Photo Parameters

NadouPro's Photo Parameters are for image generation, not general video settings. They can still help a video workflow when you first create key images, storyboard frames, character references, or product references, and then use those images in Image to Video or Multi-reference generation.

Supported Photo Parameters include Focal Length, Depth of Field, Color Science, and Lens Set. A useful order is: choose Focal Length and Depth of Field for cinematic language, choose Color Science for the base look, then choose Lens Set to define optical characteristics.

Common mistakes to avoid

• Adding too many camera movements at once: Unless you are intentionally designing a complex sequence, choose one main movement for each shot.
• Abstract action: Replace vague ideas with visible behavior, physical movement, and changes in the frame.
• Style words without anchors: Describe lighting, materials, color, lens feel, and scene details instead of relying only on labels like cinematic or premium.
• Changing character descriptions: When the same character appears again, keep the same name, clothing, proportions, and reference materials.
• Mixing settings limited to image and video workflows: Use Photo Parameters for image references, then pass the generated visuals into video generation when needed.

Recommended starting workflow

1. Write down a brief scene goal: what this shot must communicate.
2. Draft the prompt using subject, action, scene, shot, lighting, mood, references, and output intent.
3. Where visual or motion consistency is important, add NadouPro references with @ links.
4. If the prompt needs stronger cinematic phrasing, use Prompt Optimization or Blockbuster Prompts.
5. Generate a test clip in the Video node.
6. Check only one variable at a time: subject identity, action, camera movement, lighting, or rhythm.
7. Save the best references and prompt patterns as reusable materials in Workspace or Team Space.

For teams, this structure also makes collaboration easier. Directors, designers, editors, or marketers can review the same prompt sections and identify what needs to be changed instead of rewriting the entire request.

FAQ

Can NadouPro create a video from just one sentence?

Yes. NadouPro includes Blockbuster Assistant for one-sentence video generation, as well as an end-to-end workflow from script and storyboard to final edit. If you need more control, expand that sentence into structured sections before generation.

Should cinematic terms be written in technical language?

Use standard shot terms when they are clear, such as Close-Up, Low Angle, Dolly In, or Tracking Follow. If a term may be ambiguous, add a plain-English explanation of what the audience should see at the beginning and end of the shot.

How can I keep a character consistent across multiple shots?

Use stable reference materials instead of relying only on text. NadouPro supports Multi-reference input, Image Fusion, Character Body 3-View, Character Face 3-View, and Character Expression 9-Grid, all of which can be reused across multiple shots.

Can I write dialogue directly into the video prompt?

You can describe dialogue intent, but for controllable spoken performance, use NadouPro's audio tools, such as Voice convert and Text to speech; when the character's mouth movement needs to match the audio, apply Lip Sync.

Where can I try NadouPro?

You can visit NadouPro to learn more about the global platform.

Contains possibly incorrect AI-generated information. Please verify independently.

Contains AI-generated information that may be incorrect. Please verify independently.