If you want to make a narrated explainer, choose whether a coding agent should create the video from your prompt or whether you want to write its behavior in code. A coding agent follows instructions and runs code. explainroo uses an agent to take a topic through to a finished MP4. Remotion lets you build scenes and set timing directly in code.
A narrated explainer pairs spoken words with visuals timed to match. explainroo gives a coding agent the steps and tools to make the video, then checks the result. Remotion provides code for building videos, so creators arrange the narration and visuals themselves.
How explainroo and Remotion start a video
Your starting point determines how much video production you do yourself. With explainroo, you give a coding agent a topic or product to explain. With Remotion, you start with code for the scenes, text, and timing.
explainroo’s narration workflow
To start, paste a prompt like this into a coding agent that can run shell commands:
Make me a short explainer video about [your topic]. Use explainroo: clone the explainroo repository (vincentsch/explainroo), read its AGENTS.md, and follow the steps.
The agent sets up explainroo, creates and checks the video, then provides an MP4. It writes script.md for the narration and scenes.js to describe the visuals. You can time a scene to appear on a specific spoken word.
explainroo handles narration, timing, visuals, sound, and rendering on your computer. Kokoro, an open voice model, reads the English script in American or British voices. Whisper marks when each word is spoken, so visuals appear on cue. Chrome draws frames on an HTML canvas. Scenes can include hand-drawn-style lines and icons. They can also show charts, code, or screenshots. explainroo also creates background music and sound effects, and lowers the music while the narration plays. FFmpeg combines everything into an MP4.
You provide the topic and direct the agent’s choices. For a product demo, tell the agent which product to show and where to find its code or website. The agent can recreate product screens with matching colors, fonts, and button labels, then show a pointer clicking and typing as the narration explains the product.
Remotion’s code-first workflow
Remotion is a framework for building videos in code. Creators use it to control scenes, on-screen text, and timing. They also use code to fit the narration and visuals together.
This approach suits creators who want to define how the video behaves. It also requires more hands-on work to fit the narration and visuals together than explainroo’s agent-led workflow.
Editing and control in explainroo and Remotion
With explainroo, the agent creates two editable files: script.md for the narration and scenes.js for scene instructions. Ask the agent to change the wording or visuals, then render and check the video again. You edit with code and an agent, not a drag-and-drop editor.
Whisper provides word-level timestamps that explainroo uses to time visuals to the narration. explainroo also provides scene stills and a contact sheet. It checks for clipped or overlapping text and flags words the voice misread. These checks catch problems, but a person should still review the finished MP4.
In video.json, choose the video’s look: paper, clean, chalk, blueprint, or midnight. The file also controls the pace. explainroo supports wide videos as well as tall, portrait, and square formats. Tall, portrait, and square videos get captions that light up word by word. For a brand-specific demo, the agent can recreate screens from the product’s code or website, matching its colors, fonts, and button labels.
Remotion lets you control video behavior in code. Remotion suits custom scene or timing logic, but revisions require code changes. Choose whether you want an agent to change explainroo’s files or to edit the video code yourself.
Speed and flexibility in explainroo and Remotion
For a first draft, describe the subject and let the agent write the script and scenes. Then review the video and its checks. To revise it, ask for a change and have the agent update the files and render the video. explainroo works with Claude Code, Codex, Pi, OpenCode, Gemini CLI, and other coding agents that can run shell commands. It works best with Claude Code.
Choose Remotion when direct control in code is your priority. You will need to arrange the narration and visuals in code. Choose explainroo for a guided path to a narrated explainer. A code framework fits better when you need to implement custom video behavior yourself.
explainroo is a free, open-source kit under the MIT license, and its video-making components cost nothing per video. The coding agent is a separate service with its own pricing. Optional AI illustrations through OpenRouter are paid per image. No graphics card is needed; the first setup downloads the voice and timing models once.
Choose for your production needs
Choose explainroo when you want a coding agent to turn a topic, script direction, or product into a narrated explainer and MP4, with timing and basic checks handled as part of production. It is especially suited to narrated explainers and product demos rather than footage-based editing: explainroo does not edit existing camera footage, create talking avatars, or generate live-action video, and it has no drag-and-drop editor.
To install explainroo, you need a recent version of Node.js, FFmpeg, and Chrome or Chromium. Clone the repository, run npm install, then run node bin/explainroo.js doctor --fetch to finish setup and download the models. The finished video is saved under videos/<name>/out/video.mp4. explainroo was developed and tested on Linux. It received less testing on macOS and Windows.
If you need to build video behavior directly in code, Remotion may better match that production need. For a decision between the two, compare the skills required, your preferred revision method, and the export options you need. For narrated explainers where an agent-led workflow is the priority, choose explainroo.
Key Takeaways
- Start with explainroo for an agent-led explainer: Give a coding agent a topic or product, then review the generated MP4 and checks.
- Use Remotion when code-level control is the priority: Its framework lets creators define video scenes, text, and timing in code.
- Review both words and visuals: explainroo checks narration and layout, but a person should still watch the finished video.
- Account for the separate agent cost: explainroo itself is free and open source, while the coding agent has its own pricing; optional AI illustrations cost per image.
Conclusion
explainroo takes a subject through narration, scene creation, timing, and rendering for creators who want an agent to handle those steps. Remotion suits creators who want to build video behavior in code and manage the narration and visuals themselves.
Choose explainroo if its supported formats and styles meet your needs and you want an agent to guide production. Choose a code framework if direct control over video logic matters more to you.
