Qontent Tube 0.3 starts from the other end. Your real screenshots and recordings stay exactly as captured, and animation does the explaining around them.

We made our launch video this way. It's the app recording itself, on the project that produced the video. You can even see the film inside its own preview.

The animated scene didn't need an API key. Here's what we did on purpose, so you can judge it fairly:

• when there's no narration. The video uses headlines and music, and timing comes from word count. Add a voice and the timing follows it.
• The recordings are sped up 1.4× to 2.3× before import, to fit each scene.
• The recording and review were scripted so we could repeat them. That includes the on-screen pointer, which our script draws, and the privacy sign-off, since the footage shows only a demo project. The same steps work by hand in the app.
• The music is synthesized, not licensed.

Dogfood our demos​

I created the launch video using Qontent Tube 0.3 itself, capturing our real screens and animating around them. This helped us spot issues that wouldn't appear in a purely AI-generated demo.

That's why we dogfood our own demos, ensuring that what we present is faithful to reality.

Pixels you can trust​

Import a screenshot or recording, and it's verified from its bytes and hashed. Then direct it: draw named regions, move the camera, and add focus outlines and callouts on a layer above. The stage can only crop, scale, and pan with one uniform scale. It can't redraw your screen.

Each camera move is anchored to a line of dialogue, not a timestamp. Rewrite the line and the camera follows.

It would rather stop than fake it​

A broken capture doesn't quietly become something else. Validate & render links each problem to its field and offers a one-click fix only when the fix is lossless. Errors block the render, with no override.

A person signs off privacy, because we won't pretend software can catch every secret on screen. Approval is tied to hashes, so changing the source undoes it.

AI when you choose it​

Some ideas have no screen to capture. Allow AI adaptation makes a scene eligible and sends nothing. You then choose what to send, for that run only, and get a candidate to compare with your original. It renders only after you accept and approve it.

An animation is data drawn by trusted primitives, so no generated code runs, and the same seed gives the same frames. Generated results are stored, so re-rendering never calls a provider again.

If you're new to making videos​

You don't need a timeline editor, a motion designer, or a sponsor's budget. You need a script and your product on screen:

1. Write a short script in Markdown.
2. Record your screen.
3. Import it, mark what matters, and validate.
4. Render on your own computer.

Your video stays original because it's made from your work, and rendering costs nothing per minute. You pay only for the services you choose, such as a voice key. Follow the workflow.

If you're already an expert​

Everything the editor does is a diffable file that keeps your comments and key order, so it can live in your repo beside the script:

scenes:
  scene-4:
    asset: regions
    camera:
      maxZoom: 1.6
      keyframes:
        - {
            id: canvas,
            at: { segment: segment-4-0, f: 0.4 },
            region: canvas,
            zoom: 1.3,
          }

Check it in CI without starting the app. The command exits nonzero on blocking problems, and --json returns the app's own report:

qontent-tube capture validate --project launch --script tour.md --strict --json

Each render writes metadata that records what it used, with hashes and no paths or keys:

Versioning your work​

Qontent Tube keeps your scripts, storyboards, and captures versioned alongside your project. Every render writes metadata with hashes, so you can track exactly what was used and when. Reproducibility comes built in, and so does a clear history of every change.

We call this approach Video as Source: a new way of making video that we've defined, where a document or script is the source of truth and the video is a build artifact. You write the script, point it at your real captures, and the video is generated from them. Edit the source, rebuild the video.

Software teams already work this way. DevOps has Infrastructure as Code, and docs teams have Docs as Code. Video was the missing piece. With Video as Source, your script is versioned, reviewed in a pull request, checked in CI, and rendered on demand, and it always matches the product it shows.

Capture-first filmmaking is how we do it: real content comes first, and AI adaptation or animation comes after, only when you choose it.

What's next for you​

• Start with Make a product demo, then Capture scenes.
• Already on 0.2? Your scripts, storyboards, and audio keep working. Captures are optional.

Capture first. Animate everything else.

QontentTube brings those decisions into one visible workflow. You start with an idea, shape the script, direct what appears on screen, choose the voices and characters, review the result, and render the finished video—all while keeping the work connected and editable.

Start by giving the video a home​

Every QontentTube production begins with a project. A project can represent a campaign, course, topic, channel, or any other collection of related videos.

Inside that project, each video begins as a Markdown script. Markdown is simply portable text with a small amount of structure for titles, scenes, spoken lines, and visual directions. You can work inside QontentTube or edit the files with another text editor. When you return, QontentTube can detect those external changes.

As the production grows, its storyboard, audio, translations, and renders remain connected to the same project. You do not need to reconstruct where every asset belongs each time you continue working.

Write directly or begin with an assisted draft​

You can write every line yourself, or use New Draft to turn a topic or outline into a starting point.

Before generating a draft, you can choose:

• The approximate video length
• A wide or vertical format
• The roster of speakers
• A visual theme
• The starting language

Assistance is optional. You can use a connected AI service or copy QontentTube's prepared prompt into another assistant and paste the response back manually.

In either case, the result remains an editable script. You can rewrite lines, move scenes, add visual instructions, or replace the draft completely. Creating a draft does not automatically generate voices or render a video. QontentTube waits for you to review and accept the work.

Use the storyboard to direct each scene​

The script establishes what the audience hears. The storyboard establishes what they see.

A scene can include titles, cards, diagrams, images, captions, characters, and other visual building blocks. You can start from a template or generate an initial suggestion, then adjust every element yourself.

Visual timing is connected to the narration rather than fixed to a fragile timestamp. For example, you can attach the entrance of a diagram to a particular spoken line. If that line becomes longer after an edit or a regenerated voice, the visual moment moves with it.

QontentTube also separates exploration from approved work:

• Approve a scene when it communicates the right idea.
• Lock a scene when later generation should leave it alone.
• Lock individual elements when only part of the scene is final.

This makes revisions more controlled. You can improve unfinished material without replacing the decisions you already approved.

Give every speaker a voice and a character​

QontentTube treats the voice and the visible character as two related but separate choices.

A roster defines who speaks and which voice each speaker uses. A production can use one narrator or several roles. You can adjust each voice's pace and choose different voices for translated versions.

The cast defines who appears on screen. You can assign one of the built-in characters, upload an image character, or create a character from a description.

Characters are reusable across productions and come prepared for common performance states such as entering, blinking, reacting, waving, and talking. During playback, the speaking character's mouth follows the generated voice, including its natural pauses.

That reuse matters for channels and recurring content. Instead of rebuilding a presenter for every episode, you can give the audience a consistent personality they recognize.

Turn your visual identity into a reusable theme​

A theme defines the colors, typography, surfaces, captions, and accents used throughout the stage.

You can duplicate a built-in theme, change a few key choices, and preview the result immediately. Readability checks help ensure that captions and labels remain clear instead of merely looking attractive in a color picker.

Themes can apply to one script, one project, or the entire workspace. If your visual identity changes, you can update the rules instead of manually restyling every card, title, and caption.

Review the script, animation, voice, and captions together​

Script plus Stage brings the production together before rendering.

The script remains editable while the stage shows the animated result. A shared timeline aligns scenes, voices, visual moments, and captions. Clicking a scene, spoken line, or timeline block takes you directly to that point in the production.

You can preview with estimated timing before generating voice audio. Once the voices are ready, QontentTube uses their real durations so you can review the actual pace.

During playback, the current scene and spoken line are highlighted, captions appear, and the assigned character speaks. If something feels wrong, you can return directly to the word, visual, voice, or theme that needs attention.

Create translations and vertical versions without rebuilding​

Once the main production works, you can reuse it for other audiences and formats.

The Language tool creates a linked translation while preserving the scenes, speakers, and visual plan. It identifies structural differences, tracks source lines that changed, and lets you assign appropriate voices for each language. Timing checks reveal when translated narration needs adjustment.

The Mobile tool adapts the same scenes for vertical video. It can restack visual elements, divide a busy scene into two beats, enlarge captions, and show the interface areas commonly covered by platform controls.

Wide and vertical previews can appear side by side. This helps make a Short feel intentionally composed instead of looking like the middle section of a cropped landscape video.

Render predictably on your own computer​

QontentTube does not depend on a video-generation model to reinvent every scene.

Its current video engine is deterministic: the same approved instructions produce the same result. The animation is assembled from your script, storyboard, reusable characters, theme, voice timing, and captions.

That distinction makes revisions practical. Changing a word, color, or layout does not require paying a video model to imagine the entire production again.

AI can still assist with selected creative tasks, including:

• Drafting a script
• Suggesting storyboard elements
• Creating character specifications
• Translating content
• Generating an optional scene background

Voice generation uses the voice service you configure. The animation timeline and final assembly remain controlled and repeatable.

Final rendering happens on your computer. QontentTube checks that the production is ready, renders the video, and can also produce captions, chapter markers, vertical versions, and review frames. Your source files and finished renders remain local-first.

One source can become many finished videos​

The complete workflow is straightforward:

1. Create a project.
2. Write or draft the script.
3. Direct each scene in the storyboard.
4. Choose the voices and cast.
5. Apply a reusable visual theme.
6. Review the synchronized stage and timeline.
7. Adapt the production for other languages or formats.
8. Render the approved result locally.

The real value is not simply producing one video. It is keeping every creative decision connected, so revisions remain manageable and one well-structured production can serve more formats and audiences.

Start with one idea. Build one scene at a time. Let QontentTube turn the work you approve into every version you need.

Write the scenes. Hear the cast. Watch the cards slide in on the word that mentions them, the characters talk with the real voice, and the finished MP4 land in your folder — in English, in Spanish, and as Shorts.

The video below was made entirely with Qontent Tube 0.2. One Markdown script, one storyboard, two voices, a cast of two characters, three painted backgrounds, and three renders: this one, a Spanish version, and a vertical Short.

From words to video​

Studio now has three views over the same draft: Script, Script + Stage, and Storyboard. The stage is a 1920 × 1080 animated video whose timing comes from the voice. Every beat is anchored to a line — "line 2 at 50%" — never to a timestamp, so when you rewrite a sentence or regenerate a clip, the animation simply follows.

No voice yet? Press Preview with estimated timing and watch the whole video play on word-count timing.

Storyboards you can trust​

A storyboard is a JSON file next to your script — data, never generated code. Fourteen element kinds (cards, chips, flows, terminals, tables, bars, characters…) with form editors, wireframes at 50% and 90% of each scene, and a review flow: approve, lock, Approve all scenes. Edit the script, and the scenes whose lines moved are marked Needs review. Nothing silently disappears.

Draft with AI, stay in charge​

The NEW tool turns a topic into a script and a full storyboard. Bring your own Anthropic or OpenAI key (or any OpenAI-compatible server, like Ollama), choose the model and the effort, and see what to expect: Fast for first drafts in under a minute, Thorough for long, careful scripts. Or skip keys entirely: manual mode shows the exact prompt for any chat product and validates the reply you paste back.

Every result is a draft: validated against the schema and the script, repaired once automatically if needed, recorded with its provenance, and never written over your work.

Your brand on stage​

Themes give the stage your colors and fonts, graded for WCAG readability with one-click fixes. Extract a palette from your logo, apply it per script, per project, or everywhere, and export it as Design Tokens.

A cast that talks​

Build characters in the Character builder: describe them, pick a style (flat, soft 3D, line, pixel) and a personality, generate four variations, lock the ones you like, and refine them in words. Every character ships rigged — breathing, blinking, waving, expressions — and talks with the real voice: Qontent Tube reads each clip's loudness 30 times a second, and the mouth follows.

Use CAST to give every speaker a face. In the demo, our Narrator is Hapi and the Expert is Nova, built in seconds from one sentence.

Every scene its own world​

Generate background paints a picture for a scene from its story and your theme's colors, with the card areas kept calm and text kept out. A theme-colored veil keeps every word readable, and a slow zoom or pan adds life. Prefer another tool? Copy the prompt and upload your image.

One story, every language​

The LANG tool translates a script into a linked draft (intro.es-MX.md) that keeps the same scenes, speakers, and storyboard, with its own on-screen texts. Give each speaker a voice per language, protect terms with a glossary, and let Timing fit the translation to the original pace. Change the source, and only the outdated lines are re-translated.

The Spanish cut of the demo was translated, re-voiced, and rendered in a few minutes:

And every screen​

The MOBILE tool adapts each scene to 9:16 — restack, split into beats, focus crop, or letterbox — with the platform safe zones of YouTube Shorts, Reels, and TikTok drawn on the preview. The voice stays the same. Render it as one vertical video or as parts under 60 seconds, each with a "Part n of N" card.

Here is part 1 of the demo as a Short, rendered from the same storyboard and voice:

Rendered on your computer​

RENDER runs a pre-flight checklist, then exports an MP4 at 30 fps with captions (.srt), YouTube chapters, and QA frames, using the Chrome and ffmpeg you already have. Renders run in parallel chunks, can be cancelled and resumed, and audio can be re-mixed without rendering the picture again. Your scripts, voices, and videos stay on your machine.

Get started​

• Read the new guides, starting with Studio and the animated stage.
• Already using 0.1? Your scripts, rosters, and audio keep working; 0.2 only adds.

Write it. Hear it. Ship it.

The script lives in one place, voice choices somewhere else, and every paragraph becomes another file to name and place. Rewrite one sentence and the whole production starts to feel uncertain again.

Qontent Tube keeps the script at the center of that work.

Faceless videos should never feel voiceless​

Qontent Tube turns a structured Markdown script into an organized, multi-voice production. It recognizes scenes and speakers, connects those speakers to a reusable cast, estimates runtime, generates narration at the scope you choose, and plays the result as one continuous timeline.

Write the story. Cast the voices. Press play.

A script format you can keep​

The writing conventions are deliberately small:

# Why Great Ideas Never Leave the Notes App

## Scene 1 — The unfinished idea

[ON SCREEN: A phone filled with half-written notes.]

**NARRATOR:**

You probably do not need more ideas. You need a better way to finish the ones you already have.

Scene headings shape the outline. Bold speaker cues identify who talks. Bracketed directions guide the visuals without being spoken. The source remains ordinary Markdown that you can edit inside or outside Qontent Tube.

Cast once, then keep creating​

A reusable roster keeps the voice and speaking pace for every narrator or character together. A documentary narrator can remain recognizable from episode to episode. A fictional cast can stay consistent across an entire story.

The roster also makes structural problems visible before generation. If a speaker in the script does not match a configured member, Qontent Tube points it out instead of guessing.

See the production before generating it​

The workspace turns the script into a scene and speaker outline. You can jump to a scene, find every passage for one speaker, and use duration estimates to check pacing before creating audio.

Generate what changed​

The first generated draft is useful. The fifth revision is where the workflow matters most.

Qontent Tube tracks whether narration is ready, missing, failed, or outdated. If a sentence, voice, or pace changes, only the affected passage needs new audio. Unchanged clips remain ready and in sequence.

Hear one continuous story​

Generated passages become a single playback timeline. Play, pause, scrub, skip, or move between clips while the active scene and spoken source stay highlighted.

That context matters. A folder of clips tells you what was generated; a timeline tied to the script tells you how the story feels.

Built for change without chaos​

Qontent Tube began with a practical frustration: creating faceless videos with several narrators and characters meant managing too many disconnected audio files. Every small improvement to the script caused another round of searching and rebuilding.

We wanted the script to remain the heart of production, recurring voices to feel like a cast, and revision to feel safe enough that creators keep improving the story.

That is the idea behind Qontent Tube:

One script. One cast. One continuous story.

Create your first voiceover or read the Markdown script guide.