Upload a clip. Describe the change. Skrrol's video-to-video transforms existing footage with AI — style transfer, scene changes, restages — finished in a real editor.
Video-to-video is the workflow for transforming footage you already have. Instead of generating a new clip from scratch, you supply real video — a screen recording, a phone clip, a stock asset, an old AI generation — and tell the model what to change. The shape of the change can be anything from a stylistic restyling ("turn this into hand-drawn animation") to scene rewrites ("replace the background with a beach") to dressing changes ("give the speaker a suit instead of a t-shirt").
Where image-to-image is the most-used image workflow, video-to-video is on a similar trajectory in video. It's the way creators who already have footage extend their reach without re-shooting. A vlog clip becomes a stylised intro. A product walkthrough becomes a brand-styled promo. A talking-head becomes a polished version with a different setting. And because Skrrol gives you the editor on the back end, you can color-match the transformed footage to other clips, layer captions, drop in music, and export — without leaving the browser.
The prompt for video-to-video is the same shape as image-to-image: be specific about what to change and what to preserve. "Restyle this clip as a watercolour painting" is a common style transfer; "keep the speaker's face natural but stylise the background" is a more controlled version. The model returns a transformed clip at the source resolution and duration.
Video-to-video is the most credit-heavy generation on Skrrol because both the input and output are video. The Free tier covers learning the workflow; Standard at €7.99 is enough for occasional restyling; Advanced and Advanced Pro are for teams who use it as part of a regular pipeline. Every output is local-first — clips stay on your device until you export.
Convert a clip to a new style — watercolour, comic, 3D render, oil painting, pixel art, hand-drawn animation.
Replace backgrounds, change settings, swap props — while keeping the subject's motion intact.
Change a speaker's clothing, swap product variants, recolour elements without re-shooting.
Outputs match the source resolution and duration where the model supports it.
Color match, add transitions, layer captions and music in Skrrol's editor on the same canvas.
Source clips and outputs live in OPFS on your device. No content cloud.
Result: A stylised version of the original clip — useful for branded content, explainer videos, and music-video aesthetics.
Result: A polished studio-look version of a phone-shot walkthrough — saves a real studio booking.
Result: A new wardrobe pass on existing footage — useful for repurposing old clips into corporate-feeling content.
Result: A modern-looking version of an old clip suitable for re-publishing in a documentary or retrospective edit.
Sign in, click Generate, pick the Video tab, and select Video-to-Video.
Drop in a video clip from your device. Source resolution and aspect ratio drive the output canvas.
Be specific about what to change and what to keep. "Restyle as watercolour, keep speaker's face natural" is more reliable than "make it artistic".
Hit Generate. The transformed clip appears in your library when ready.
Color match to the rest of your footage, add captions and music, trim, and arrange on the timeline.
Render to MP4 in your chosen aspect ratio, ready to publish.
Skrrol AI uses VL credits across all generators — image, video, voice, and music. The same credit pool applies; heavier modalities (video) use more credits per generation than lighter ones (image, voice). Choose a plan and use credits across any generator.
Trial credits to generate a few short test clips. a few hundred to a thousand credits per clip on the higher-quality models, less for fast/preview tiers. Watermarked.
8000 VL credits — roughly the volume needed for a steady weekly upload schedule when each clip costs about a few hundred to a thousand credits. No watermark.
17000 VL credits — comfortable headroom for daily iteration, B-roll generation, and ad variants without rationing prompts.
35000 VL credits — production volume for studios shipping multiple finished video pieces per week and re-rolling generations to dial in shots.
Each model has its own cap. Skrrol surfaces the maximum at the top of the panel. For longer pieces, split the source into segments and re-stitch on the timeline.
Yes if the prompt asks for it ("keep the face natural"). Aggressive style transfer can drift the identity — re-roll with a more controlled prompt if it does.
Video-to-video is the most expensive generation on Skrrol because both input and output are video. Standard (€7.99) is fine for occasional use; Advanced and Advanced Pro are for regular pipelines.
Yes — Skrrol can carry the source audio through unchanged, or you can replace it with AI-generated voiceover and music from the editor.
Stylistic restyling, background swaps, and wardrobe / detail edits are the strongest cases. Extreme structural changes (different actions, different shots) are better handled by re-generating from text-to-video.
The source goes to the underlying model provider so the model can run. Skrrol doesn't retain it on a content cloud — outputs land in your local project.
Every generation opens directly in the Skrrol editor. These features are particularly useful as the next step after a video-to-video ai generator run.
Skrrol AI runs every generator next to a full pro editor. Your work stays on your device. Start free.