Appearance

Account Settings Sign Out
AI generation is temporarily paused. Image, video, and audio generation are disabled by an administrator. Existing content is unaffected.
Docs Home /Scene Mode /Overview

Scene Mode

Scene Mode is the script-to-video surface. Paste a script, click Generate, and Fawna chunks it into scenes, generates an image for each, adds narration, and produces a finished video. When the output is close, you can hand it to the Editor for fine control. Minimum effort, maximum pacing.

When to use Scene Mode

  • You already have a script and want a video with a narrator reading it.
  • You care about pacing and narrative flow more than shot-by-shot camera work.
  • You want a first-draft video in minutes, not hours.
  • You are producing educational, explainer, or voice-driven content where each beat is a scene.

The flow at a glance

  1. Create a storyboard. Paste your script, pick an aspect ratio, pick a voice (or a cast of voices). See Creating a storyboard.
  2. Review the scene split. Fawna breaks the script into scenes based on pacing. Each scene gets its own card in a grid.
  3. Generate assets. Bulk-generate images, narration audio, and motion video for every scene. Re-roll individual scenes until each one lands.
  4. Play it back. The storyboard player stitches every scene with its audio. Preview from start to finish.
  5. Import to the editor if you want to cut, add transitions, layer audio, or add overlays. See Importing to the editor.

The workspace

Scene Mode opens with a preview player on top and a scene grid underneath. The sidebar on the right toggles between two contexts:

  • Storyboard panel: global settings, script, character list, model defaults.
  • Scene panel: edits for the scene you have currently selected. Text, image tier, motion model, speaker, placement.

Click any scene card in the grid to focus it. The preview scrubs to that scene and the sidebar switches to the scene panel.

Scenes are the unit of work

A scene is one beat of the script. It has a text chunk, an image, an audio clip of the narrator reading the chunk, and optionally a motion video generated from the image. Every scene is independent: regenerating one does not affect the others.

Fawna picks scene length based on scenes-per-minute, a pacing dial. Default is 10, which gives roughly 6 seconds of finished video per scene. You can adjust this at storyboard creation.

Scene Mode vs Editor

NeedUse
Script to finished video fastScene Mode
Multiple overlapping audio tracksEditor
Cross-fades and transitionsEditor
Text overlays, lower thirdsEditor
Specific trim down to frameEditor
Different aspect per sceneNeither (one aspect per storyboard)
Fine shot-by-shot controlEditor
30-scene narrative in 20 minScene Mode, then import to Editor for polish

Typical workflow: draft in Scene Mode, import to the Editor, polish there. Scene Mode handles the coarse shape, Editor handles the detail. The two are decoupled: edits you make in the Editor do not push back to Scene Mode.

Where to go next

Storyboard
Scene
Replace a shot, or insert a new one