explainroo makes explainer videos and product demos with narration, diagrams, and product screens, without an on-screen presenter. HeyGen focuses on avatar-led videos.
explainroo is a free, open-source kit that works with an AI coding agent - a command-line tool that writes and runs code from your instructions. Describe the video, and the agent creates it on your computer. You control the scenes and narration, but there is no drag-and-drop editor.
Avatar-led vs. avatar-free explainer videos
The main choice is whether a visible speaker should lead the explanation. An avatar gives viewers a presenter to watch, which suits messages delivered directly to camera. For topics better explained with diagrams, interface visuals, or charts, an avatar-free format keeps the focus on the subject.
HeyGen’s avatar-led format
HeyGen makes videos featuring on-screen avatars. A visible presenter suits an announcement or introduction led by a speaker. If you want visuals to show a process, product interface, or set of ideas while narration explains them, an avatar may not fit. Check HeyGen’s current product information for its options and pricing.
How explainroo makes avatar-free explainers
explainroo builds scenes around spoken narration. Kokoro, an open voice model, reads the script in natural American or British English. Whisper, a speech-recognition tool, records when each word is spoken so scene elements appear on cue.
Chrome draws the frames on an HTML canvas, the area where a browser can render graphics. Scenes can use hand-drawn-style lines and icons. They can also include charts, code, and screenshots. explainroo also makes background music and sound effects, lowers the music while narration plays, and combines the finished video into an MP4 file.
To make a video, paste this prompt into a coding agent that can run shell commands, replacing the bracketed words with your topic:
Make me a short explainer video about [your topic]. Use explainroo for it: clone https://github.com/vincentsch/explainroo, read its AGENTS.md and follow the steps.The agent sets up explainroo, creates the video, checks it, and gives you the MP4. It writes script.md, which contains the narration, and scenes.js, JavaScript instructions for drawing the scenes. explainroo handles the rest on your computer. Claude Code works best with explainroo. It also supports Codex, Pi, OpenCode, Gemini CLI, and other coding agents that can run shell commands.
The agent checks scene stills and a contact sheet of frames. It also checks for cut-off or overlapping text and mispronounced words. Choose a visual style in video.json, such as paper, clean, chalk, blueprint, or midnight. The file also controls pace. In tall, portrait, and square formats, captions light up word by word.
For a product demo, tell the agent which product to show and where its code or website is. explainroo can rebuild screens using the product’s colors, fonts, and button labels, then show a pointer clicking and typing while narration explains the interface.
explainroo’s fit, limits, and costs
explainroo suits creators who want the script and visuals to carry the explanation. It can show a sequence of concepts or highlight a feature in a screenshot. It can also guide viewers through product screens without a presenter on screen. Formats include wide video for YouTube and tall video for short-form platforms, along with portrait and square video.
explainroo has no drag-and-drop editor. You describe changes to the coding agent, which edits the script or scene instructions. It also does not edit existing camera footage or create talking avatars or AI-generated live-action footage. Choose another format if those are central to the project.
The kit is free and released under the MIT open-source license. Optional AI illustrations through OpenRouter cost money per image, and the coding agent is a separate service with its own pricing. Everything else in explainroo costs nothing per video. The video kit and the coding agent have separate costs.
Choose explainroo or an avatar-led format
Choose explainroo when you want narration paired with diagrams, screenshots, code, charts, or recreated product screens. To test the fit, give the agent one focused topic and review the MP4, scene stills, and layout check before expanding the project.
Choose an avatar-led format when a visible speaker is central to the message. Check HeyGen’s current features and pricing before you commit. Use an avatar-oriented approach when the video needs a person on screen. Use explainroo when the subject and visuals should lead.
explainroo vs. avatar-led video: key takeaways
- Choose the format first: Use explainroo when narration and supporting visuals should lead. Choose an avatar-led approach if a visible presenter matters.
- Try explainroo with a focused prompt: Give a coding agent your topic, and it creates the script, scenes, checks, and MP4.
- Plan for agent-based editing: You request changes through a coding agent rather than editing scenes in a drag-and-drop interface.
- Separate kit costs from service costs: explainroo is free under the MIT license, while the coding agent has separate pricing and optional AI illustrations cost per image.
Conclusion
explainroo suits explainer videos that use narration, timed visuals, and product demos instead of avatars. A coding agent on your computer creates the video and runs automated checks.
Choose explainroo when visuals should carry the explanation and you are comfortable describing edits to an agent rather than moving elements in a visual editor.
