Image Generation
For each slide in a video, the platform generates a unique scene image that serves as the visual backdrop for that slide’s narration. Images are created in parallel for all slides, so the overall video generation time doesn’t increase proportionally with the number of slides.
How Images are Composed
Section titled “How Images are Composed”Each image is shaped by three inputs:
- The slide’s content — the factual information from your documents that the slide covers.
- The visual style you selected when launching the video job.
- Your steering prompt, if you provided one — this can influence both the content focus and the visual direction.
The platform composes a visual description for each slide and passes it to the image generator. You don’t write these descriptions yourself; they are created automatically based on your content and style choices.
Visual Styles
Section titled “Visual Styles”The visual style you select when launching a Video job shapes how every image looks:
| Style | Effect on Images |
|---|---|
| Classic | Professional corporate look; clean backgrounds; minimal design |
| Dark | Dark backgrounds; high contrast; premium feel |
| Whiteboard | Hand-drawn illustration; sketch aesthetic |
| Neon | Vibrant colours; tech-forward aesthetic |
| Watercolor | Soft, painterly art style |
Aspect Ratio
Section titled “Aspect Ratio”All video images are generated in 16:9 format, the standard widescreen ratio, so they fit the video frame cleanly.
What Happens if an Image Fails?
Section titled “What Happens if an Image Fails?”If the platform cannot generate an image for a particular slide — for example, due to a content policy restriction — that slide is skipped and the video is assembled from the remaining slides. As long as at least one slide can be completed, the overall video job will still succeed.
Advanced Usage Notes
Section titled “Advanced Usage Notes”Use these patterns to get more predictable results:
- Keep inputs specific and scoped to one intent per generation request.
- Save intermediate outputs often, especially before running major edits or regenerations.
- Use review and assignment workflows early to reduce rework late in the lifecycle.
Troubleshooting
Section titled “Troubleshooting”If results are not what you expected:
- Verify source quality first (missing context, inconsistent structure, or conflicting inputs reduce output quality).
- Re-check selected options and audience/profile settings before re-running.
- If a job fails repeatedly, wait briefly and retry once; then escalate with the artifact or job context.
Frequently Asked Questions
Section titled “Frequently Asked Questions”Why did quality change between runs?
Section titled “Why did quality change between runs?”Different input context, settings, and model/runtime conditions can produce different outputs.
Should I regenerate or manually edit?
Section titled “Should I regenerate or manually edit?”Regenerate when the core structure is wrong; manually edit when only wording, ordering, or style needs adjustment.
What is the best way to reduce iteration cycles?
Section titled “What is the best way to reduce iteration cycles?”Start with clear scope, run smaller batches, and use review checkpoints after each major step.