9. AI generated images and videos#

Sometimes an agent gives you the listing but no twilight exterior shot — and you need one for the closing scene. JSON2Video can generate b-roll images and short videos with text-to-image and text-to-video models, all from the same image and video elements. This chapter adds a twilight AI exterior shot and a short AI b-roll clip.

Prerequisites: chapter 8. You should know the difference between scene-level and movie-level elements.

Step 1 — AI images#

Instead of src, supply a prompt and a model on an image element. JSON2Video pipes the prompt to the chosen model and embeds the result.

{
  "type": "image",
  "prompt": "A craftsman house at twilight with warm lights, magazine real-estate photo, cinematic, high detail",
  "model": "flux-schnell",
  "aspect-ratio": "horizontal",
  "duration": 4
}

The full list of image models — with their IDs, output sizes, and per-model parameters — lives in the live AI models catalog. This tutorial uses flux-schnell as a fast, low-cost default; swap the model field to try others.

aspect-ratio accepts horizontal, vertical, or square. The renderer picks compatible model output sizes.

Step 2 — AI videos#

Short AI clips work the same way on a video element:

{
  "type": "video",
  "prompt": "Slow dolly shot through a sunlit modern kitchen, real-estate listing style",
  "model": "kling-1-6",
  "aspect-ratio": "horizontal",
  "duration": 5
}

AI video generation is the most credit-expensive operation in the catalog — check Credit consumption before running this at scale.

Step 3 — Add a closing twilight scene#

Append a fourth room scene whose photo is AI-generated. The output stays cached after the first render, so future renders reuse the same image at no extra cost (chapter 16 covers caching).

{
  "duration": 4,
  "transition": { "style": "fade", "duration": 0.5 },
  "elements": [
    {
      "type": "image",
      "prompt": "A craftsman house at twilight with warm lights, magazine real-estate photo, cinematic",
      "model": "flux-schnell",
      "aspect-ratio": "horizontal"
    },
    { "type": "text", "text": "Twilight Exterior", "position": "top-left", "x": 60, "y": 60 }
  ]
}

Tip — flux-schnell returns in ~5 s. flux-pro and AI video generators can take 30–60 s; expect the overall render to slow down. Always pair AI generation with cache: true (the default) so subsequent renders of the same prompt are instant.

The complete final JSON#

{
  "resolution": "full-hd",
  "elements": [
    {
      "type": "audio",
      "src": "https://cdn.json2video.com/assets/audios/uplifting-corporate.mp3",
      "volume": 0.4
    },
    {
      "type": "voice",
      "text": "Welcome to 123 Oak Street. Take a tour of the kitchen, bedrooms, and backyard.",
      "voice": "en-US-EmmaMultilingualNeural",
      "start": 1.5
    },
    {
      "type": "subtitles",
      "language": "en",
      "settings": {
        "style": "boxed-word",
        "font-family": "Inter",
        "font-size": 90,
        "font-color": "#FFFFFF",
        "outline-color": "#000000",
        "position": "bottom-center",
        "all-caps": true,
        "box-color": "#0E7C66"
      }
    }
  ],
  "scenes": [
    {
      "duration": 4,
      "elements": [
        {
          "type": "component",
          "component": "basic/000",
          "settings": { "headline": "FOR SALE", "subline": "123 Oak Street" }
        }
      ]
    },
    {
      "duration": 4,
      "transition": { "style": "fade", "duration": 0.5 },
      "elements": [
        { "type": "image", "src": "https://cdn.json2video.com/assets/images/sample-house-front.jpg" },
        { "type": "text", "text": "Exterior", "position": "top-left", "x": 60, "y": 60 }
      ]
    },
    {
      "duration": 4,
      "transition": { "style": "fade", "duration": 0.5 },
      "elements": [
        { "type": "image", "src": "https://cdn.json2video.com/assets/images/sample-house-kitchen.jpg" },
        { "type": "text", "text": "Chef's Kitchen", "position": "top-left", "x": 60, "y": 60 }
      ]
    },
    {
      "duration": 4,
      "transition": { "style": "fade", "duration": 0.5 },
      "elements": [
        { "type": "image", "src": "https://cdn.json2video.com/assets/images/sample-house-bedroom.jpg" },
        { "type": "text", "text": "Master Bedroom", "position": "top-left", "x": 60, "y": 60 }
      ]
    },
    {
      "duration": 4,
      "transition": { "style": "fade", "duration": 0.5 },
      "elements": [
        {
          "type": "image",
          "prompt": "A craftsman house at twilight with warm lights, magazine real-estate photo, cinematic",
          "model": "flux-schnell",
          "aspect-ratio": "horizontal"
        },
        { "type": "text", "text": "Twilight Exterior", "position": "top-left", "x": 60, "y": 60 }
      ]
    }
  ]
}

Expected output#

A 20-second listing that ends on a generated twilight exterior of the house. The voice-over has been updated to invite the viewer on a tour, and the subtitles transcribe the new line. Sample render: tutorial-09.mp4 (placeholder).

What you learned#

• Swap src for prompt + model to ask JSON2Video to generate the image or video.
• aspect-ratio selects compatible output sizes.
• AI generation is cached by default — repeat renders are instant and free.
• AI video is the most credit-expensive primitive; prefer AI images where possible.

Previous chapter / Next chapter#

← 8. Automatic subtitles · 10. Variables →