A Codex Skill that turns a Chinese spoken line or abstract idea into pen-line sketch explainer animation assets with selective colored-pencil fill, generated via Gemini.
28739PythonMITUpdated 3 months ago
hi-nikola/hand-drawn-explainer-video-nikola
A Codex Skill that turns Chinese explainer scripts or SRT files into hand-drawn videos with stroke-by-stroke drawing, plus program-driven animation.
- Stars
- 410
- Forks
- 59
- Open issues
- 1
- Last commit
- 2 months ago
What is it?
This repository provides a Codex Skill for producing Chinese knowledge-explainer videos from a topic, a script, or an SRT file. It offers two production routes: stroke-by-stroke story animation, where a hand-drawn canvas fills in region by region alongside narration, and programmatic animation built from HTML, SVG and GSAP for flowcharts, relationship diagrams and precise on-screen text. A prompt-only mode outputs self-contained prompts for other image or video models.
The stroke-story route supports single-scene, multi-scene and left-right semantic-island layouts, with a Q-style character look or a minimal black-character style. Outputs include a real MP4, subtitles, a timeline and editable assets. Final delivery checks the video, audio track, subtitles, scene boundaries and last frame. Voiceover defaults to a Volcengine TTS voice, and a paid external service is needed for new narration or generated illustrations.
Core stroke rendering runs locally with traditional image algorithms and needs no GPU or downloaded neural models when suitable source art and narration already exist. FFmpeg, Node.js and HyperFrames are required for full MP4 output and program animation.
Key features
- 01Stroke-by-stroke drawing that builds the picture as narration plays.
- 02Left-right semantic island layout that keeps earlier content on screen.
- 03Q-style character and minimal black-character visual styles.
- 04Programmatic HTML/SVG/GSAP animation for flowcharts and diagrams.
- 05Outputs real MP4, subtitles, timeline and editable project assets.
Use cases
- Explaining a historical figure's biography as a hand-drawn story video.
- Turning a Chinese script or SRT file into a whiteboard-style explainer.
- Producing flowcharts or concept diagrams with precise labels and arrows.
- Generating prompts for external image or video models without rendering.
Quick start
- 1
Clone the repo into your Codex skills folder
git clone https://github.com/hi-nikola/hand-drawn-explainer-video-nikola.git "$env:USERPROFILE/.codex/skills/hand-drawn-explainer-video-nikola" - 2
Run the setup check script
python "$env:USERPROFILE/.codex/skills/hand-drawn-explainer-video-nikola/scripts/setup_check.py" - 3
Prepare the stroke-video runtime environment (needed for stroke-by-stroke videos)
python vendor/srt-whiteboard-animation/scripts/prepare_env.py
Things to know
- Full MP4 output requires FFmpeg/FFprobe, and program animations require Node.js, a browser and HyperFrames.
- Voiceover defaults to a Volcano Engine voice, which needs an authorized account; system or low-quality TTS is not recommended as a substitute.
- Generating new illustrations or voiceover depends on external services that may charge fees.
Topics
FAQ
- What is hand-drawn-explainer-video-nikola?
- A Codex Skill that turns Chinese explainer scripts or SRT files into hand-drawn videos with stroke-by-stroke drawing, plus program-driven animation. This repository provides a Codex Skill for producing Chinese knowledge-explainer videos from a topic, a script, or an SRT file. It offers two production routes: stroke-by-stroke story animation, where a hand-drawn canvas fills in region by region alongside narration, and programmatic animation built from HTML, SVG and GSAP for flowcharts, relationship diagrams and precise on-screen text. A prompt-only mode outputs self-contained prompts for other image or video models.
- Is hand-drawn-explainer-video-nikola free to use? What license does it have?
- hand-drawn-explainer-video-nikola is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is hand-drawn-explainer-video-nikola still maintained?
- Updated within 6 months. The last commit to hand-drawn-explainer-video-nikola was on Sep 3, 2026.
- What can I use hand-drawn-explainer-video-nikola for?
- Explaining a historical figure's biography as a hand-drawn story video. Turning a Chinese script or SRT file into a whiteboard-style explainer. Producing flowcharts or concept diagrams with precise labels and arrows. Generating prompts for external image or video models without rendering.
- How do I install hand-drawn-explainer-video-nikola?
- Clone the repo into your Codex skills folder: git clone https://github.com/hi-nikola/hand-drawn-explainer-video-nikola.git "$env:USERPROFILE/.codex/skills/hand-drawn-explainer-video-nikola" Run the setup check script: python "$env:USERPROFILE/.codex/skills/hand-drawn-explainer-video-nikola/scripts/setup_check.py" Prepare the stroke-video runtime environment (needed for stroke-by-stroke videos): python vendor/srt-whiteboard-animation/scripts/prepare_env.py
- Where can I get hand-drawn-explainer-video-nikola?
- The source code and installation instructions are on GitHub: https://github.com/hi-nikola/hand-drawn-explainer-video-nikola
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A Codex Skill that guides AI agents to generate white-background, hand-drawn 16:9 illustrations for Chinese articles, featuring the Xiaohei character.
12.7K1.5KMITUpdated 26 days ago
Open-source speech recognition toolkit for offline and streaming ASR, with VAD, punctuation, speaker labels and OpenAI-compatible serving.
20.8K2.1KPythonMITUpdated 2 days ago
A 0.1B-parameter multilingual text-to-speech model that runs in realtime on CPU, with voice cloning and 48 kHz stereo output.
4.5K562PythonApache-2.0Updated last month
Translates scientific PDF papers into another language while preserving formulas, charts, tables and layout, with a bilingual output option.
37.7K3.4KPythonAGPL-3.0Updated 13 hours ago
Free online chapter-by-chapter companion for Qiu Xipeng's Chinese deep learning textbook, in a full edition and a general-reader edition, with PDFs and errata.
19.5K3.7KNo licenseUpdated last month