Skip to content
Vidsly

Script to video

Script in. Animated full story out.

Paste a script and Vidsly returns a finished MP4: the script cut into scenes, an AI image and a motion clip generated for every scene, narration in the voice you pick and an optional music bed — as a 9:16 Short up to 3 minutes or a 16:9 long-form video.

0 credits / clip9:1616:9Shorts to 3:00MP4
Paste a scriptSee the plans

Full videos on Creator, Pro and Studio · you see the scene count and runtime before anything renders

9:16
One scene, generated for the line of narration it sits under.
0
credits per clip on the unlimited lane
from $29 / mo
3:00
longest 9:16 Short
up to 34 scenes
7:00
longest 16:9 video in practice
70 scenes × 6s clips
20
clips a day on Creator
one scene is one clip
01What you paste

Any script. Even a messy one.

Paste the narration you want read, or give the writer a topic and let it draft one in any of 15 languages. A clean script works; so does a production document with notes in it — the Director pulls out the narration and the people and places, and leaves the rest.

Scene numbers in your text are read as text, not instructions. Scenes are cut where the picture has to change: a new place, someone entering, a jump in time, a turn in tone.

Call sheet — before anything renders

Scenes
cut from your script’s beats
Runtime
the length of the narration
Credits
0 on the unlimited lane
Renders
only when you confirm

The Studio shows you this after it reads your script. You can lower the scene count there — neighbouring beats merge into longer shots and the narration stays the same — or go back and edit the script. Nothing renders until you confirm.

02One pass

The script is the clock.

Five stages in one render. The narration sets every cut, so the picture never runs ahead of the voice or lags behind it.

  1. 01

    Read

    The Director reads the whole script before it draws anything and writes one locked description for every person and recurring place, so the last scene is drawn from the same notes as the first.

  2. 02

    Cut

    The narration is split into scene-sized lines. Each clip lasts as long as its line takes to say — up to 6 seconds on the unlimited lane — and a longer line becomes two scenes instead of a frozen shot.

  3. 03

    Voice

    Pick a narrator from 51 voices in the main picker plus 34 more in the full library (85 in all), across 16 languages. It reads your words line by line over the cut.

  4. 04

    Picture

    One AI image per scene in one of 6 styles (Cinematic Photo by default), then a planned camera move turns each still into motion that reads as one film.

  5. 05MP4

    Deliver

    One of 5 music beds mixed low under the voice, or none, then a single H.264 / AAC MP4 at 24fps. We email you when it is ready.

03Two formats

One script, two shapes.

Each with a real ceiling — and where the practical limit is lower than the format’s, we print both.

Shorts

9:16
Canvas
1080×1920
Runtime
up to 3:00
Scenes
1 – 34
Posts to
YouTube Shorts, TikTok, Reels

3 minutes is the Shorts and Reels ceiling. A script that would read longer than that is stopped before planning, with a prompt to switch to 16:9 — it is never cut short to fit.

Long-form

16:9
Canvas
1920×1080
Format bound
up to 10:00
In practice
about 7:00
Scenes
1 – 70

The 16:9 format allows 10 minutes, but full videos render on the unlimited lane, where a clip is at most 6 seconds — so 70 scenes is about 7 minutes. A longer script is best published in parts.

Make a ShortMake a long-form video

Deeper on each: the 9:16 shorts generator, long-form documentaries and one character across every scene.

04Trade-offs

What it won’t do, said up front.

Four things people expect from a script-to-video tool that this one does differently or not at all.

No captions

There is no subtitle step on the AI pipeline and no .srt file. Add captions in your uploader or editor. Burned-in captions exist only in the photo-video tool, where you bring your own images and audio.

No words in the picture

Scene images are generated without text — no titles, signs, dates or labels. Title cards and lower-thirds go on in your editor.

Softer than native HD

On the unlimited lane the 1080×1920 or 1920×1080 file is scaled up from a 480p source (720p on Unlimited Premiere, Pro and Studio; 1080p on Unlimited Ultra, Studio), so it looks softer than a native render.

A daily allowance, not a firehose

20 clips a day on Creator, 40 on Pro, 60 on Studio, resetting at midnight UTC — and renders are not instant, so plan a day ahead.

A video needing more clips than the day has left starts now with what fits and finishes automatically after the midnight-UTC reset — over up to 5 days for a very long one. You never re-submit it, and it never costs credits. How the unlimited lane works.

05Questions

Before you paste your first script.

How do I turn a script into a video with Vidsly?
Open the Studio, choose Video, paste your script and pick a format — a 9:16 Short or a 16:9 long-form video — plus a voice, an image style and, if you want one, a music bed. The Director reads the script and comes back with a call sheet: how many scenes it became, how long the video will run and what it costs. Nothing renders until you confirm. When the file is done you get an email, and you download one MP4 with the narration, scenes, motion and music already mixed.
Will Vidsly read my script word for word?
Close, not exact. The Director keeps your plot, names, facts and ending, but it tightens wording to fit the scene pacing, so expect some sentences to come back shorter. It never adds events or changes outcomes, and splits a long sentence across two scenes when one clip cannot cover it. Scene numbers or “one paragraph per scene” notes in the text are treated as content, not instructions — scenes are cut where the picture has to change. If exact wording matters, such as a quote or a disclaimer, listen back before you publish.
How long can a script-to-video be?
A 9:16 Short runs up to 3 minutes across up to 34 scenes; a script that reads longer is stopped before planning and pointed at 16:9. The 16:9 format allows up to 10 minutes and 70 scenes, but full videos render on the unlimited lane, where a clip is at most 6 seconds — so about 7 minutes is the practical ceiling for one long-form video. Longer scripts are best published as parts. Every scene is one clip from your daily allowance, so on Creator a long video can take more than a day; it finishes on its own after the reset.
What does it cost?
Full script-to-video runs on the unlimited lane, which costs 0 credits per clip — scene images included — on Creator, Pro and Studio (Creator is $29 a month, Pro $49, Studio $99). What bounds it is a daily clip allowance rather than a bill. A free account (150 credits to start, no card) can try the script writer and the voices, but full videos start on Creator.
How many videos can I make a day?
One scene is one clip, and the unlimited lane allows 20 clips a day on Creator, 40 on Pro and 60 on Studio, resetting at midnight UTC. At up to 6 seconds a clip, that is at most 2:00 of new footage a day on Creator, 4:00 on Pro and 6:00 on Studio. HD clips spend more of the same allowance: a 720p clip on Unlimited Premiere counts 2× and a 1080p clip on Unlimited Ultra counts 4×. A video needing more clips than the day has left starts now with what fits and finishes automatically after the midnight-UTC reset — over up to 5 days for a very long one. You never re-submit it, and it never costs credits.
What resolution is the finished video?
The file is 1080×1920 for a Short and 1920×1080 for long-form, but on the unlimited lane that canvas is scaled up from a 480p source (720p on Unlimited Premiere, Pro and Studio; 1080p on Unlimited Ultra, Studio). It looks softer than a native render, and we would rather say so here than have you notice it at upload. When one shot has to be sharp, the per-clip engines in the Studio's single-scene mode render native 1080p from your monthly credits.
Does it add captions or on-screen text?
No. The AI video pipeline has no subtitle step and delivers no .srt file, and scene images are generated without words in them, so captions, title cards and lower-thirds go on in your editor or your uploader. Burned-in captions exist only in the photo-video tool, where you supply your own images and audio.
Can I keep the same character in every scene?
Yes. Pick a saved character, describe one, or upload a reference image, and every scene image is drawn from that one character reference rather than from the previous frame — so the person in the last scene is the person in the first. Recognizable copyrighted characters are refused before you start, and photos of real people are often refused by the image model's safety check, so an original character or a drawing works best.
Can the AI write the script for me?
Yes. Give the Studio's writer a topic and it drafts a script in any of 15 languages, which you can edit before generating. Drafts are counted per month: 100 on Creator, 200 on Pro and 200 on Studio.
How long does a video take to render?
Long enough that you should not wait on the page. Every scene is its own image and its own motion clip, so a long video takes longer than a Short, and the unlimited lane is not built for speed. You get an email when the MP4 is ready; we do not quote a per-clip time we cannot stand behind.
Related

Keep exploring

The script is the hard part. The rest is one pass.

0 credits per clip on the unlimited lane — 20 clips a day on Creator from $29/mo, resetting at midnight UTC.

Paste a scriptSee the plans