Turn your Python code into video walkthroughs for your docs
Point Orange Brackets at a Python script or a Jupyter notebook. It runs the code for real, writes a script grounded in what actually happened, records a real screencast, and narrates it, ready to drop straight into your docs, changelog, or onboarding guide.
60 free credits on sign-up · no credit card required
How it works
Four stages, from code to finished MP4
Everything up to reading the script is free, so nothing is spent before a human approves the words the video will actually speak.
Settings
Pick a workspace and a notebook or script, then set language, voice, theme, and resolution. Submitting runs the code for real in a sandbox.
Scripts
Narration is drafted from what the code actually printed, editable line by line. The exact credit cost is shown before anything is spent.
Audio
Every line is voiced by the chosen speaker, each with its own regenerate. Rendering then captures a real screencast timed to the narration.
Complete
The finished MP4 plays in-page and downloads directly. Rendering is never metered twice; a failed render can always be retried free.
See it live
This is what your video is captured from
A real instance of the same sandbox environment every video renders from -- pick a format and a theme to see exactly how it looks.
Theme
Dark
Light
All 10 editor themes are available for your own videos -- this is the exact same rendering pipeline.
What makes it different
Not slides with a voiceover: a real recording of real code
A real screen recording
The final video is captured from an actual running browser, not composited from slides; the code really types, runs, and produces the output shown.
Narration grounded in real output
The script reacts to what the sandbox run actually printed, rather than a generic description of what the code is expected to do.
Camera that follows the code
Notebook videos get a deliberate zoom into the cell and its output as it's discussed, then pull back for narration, timed automatically.
Your brand name, said right
Every workspace has its own pronunciation rules: a respelling or phonetic override for the product names and acronyms a voice model would otherwise guess wrong, applied at narration time, before a single credit is spent regenerating it.
A workspace remembers its script
A new project from a notebook already scripted for this workspace skips straight to review, with no re-running the sandbox and no rewriting from scratch.
Five languages, curated voices
English, Spanish, French, Arabic, and German, each with its own paired, previewable speaker that auto-adjusts if the language changes.
Video Formats
One pipeline, three formats
Single Video
One narrated screencast from a notebook or script: the full tutorial format, landscape, up to 4K.
.py / .ipynb · 720p–4KShort Videos
A 30-60 second vertical short from a .py file, fixed at 1080×1920 with burned-in captions, built for feed placements.
.py only · 1080×1920.py Scripts
The same narrated pipeline, sourced from a plain Python script instead of a notebook: one continuous, line-numbered editor view, typed and run for real.
.py · 720p–4KPricing
Only narration is metered
Running the code, writing the script, and rendering the final video are always free. Text-to-speech is billed at roughly 48 characters per credit, about 20 credits per finished minute of video. Every pack unlocks the exact same features — every language, every voice, every theme, every resolution. Bigger packs just mean more credits, never more capability.
Free
$0
60 credits, one-time
About 3 minutes of finished narrated video. Enough to try the whole pipeline before deciding on anything.
X-Small
$12.99
300 credits (~15 min)
About 15 minutes of finished narrated video. For a quick top-up smaller than a full pack.
Small
$25.99
600 credits (~30 min)
About 30 minutes of finished narrated video. Enough for a handful of tutorials, or to top up between free runs.
Medium
$63.99
1,500 credits (~75 min)
About 75 minutes of finished narrated video. Enough for a full multi-part course.
Large
$191.99
4,500 credits (~225 min)
About 225 minutes of finished narrated video. Enough for a full course catalog.
Beta pricing: credits only, no subscription yet. Buy another pack anytime you need more.
FAQ
Frequently asked questions
Orange Brackets turns your Python code into a narrated video walkthrough for documentation and tutorials. It runs the code for real, writes a narration script grounded in the actual output, records a real screencast, and voices it in a chosen language, with no editing software or manual voiceover.
Your first video is free
60 credits, no credit card. Connect a script or notebook and see the whole pipeline run once before deciding on anything.
Get started free