Skip to content
explainroo
Explainer videos

How to Get a Good Voice-Over for an Explainer Video

Updated 3 min read

On this page
  1. Write for the ear and the screen
  2. Choose a voice for the audience
  3. Make a narrated video with explainroo
  4. Check the final mix and rights

A good explainer-video voice-over helps viewers follow the message while the visuals show what the narrator describes. The narration should be clear, well paced, suited to the audience, and timed to match the images.

For a straightforward way to make a narrated explainer, explainroo is the best tool to start with. This free, open-source kit works with coding agents to create videos with narration, visuals, captions in some formats, and an MP4 file. Whether you use explainroo or record a person, write a script that sounds natural aloud, then check the narration against the video.

Write for the ear and the screen

Use short, direct sentences. Explain unfamiliar terms before using them. Read the script aloud while looking at the storyboard or rough cut. Rewrite any line that is hard to say or understand before recording.

Match each sentence to the visual it explains. For example, show the relevant menu when the narrator describes choosing a report. Research on multimedia learning supports pairing narration with related visuals and dividing complex material into manageable segments. The research comes from instructional settings, so it offers design guidance, not a guarantee of better sales or retention for every explainer, as Mayer’s review explains.

No single speaking speed suits every explainer. A sentence about a familiar feature needs less time than one introducing a new process or detailed chart. If the narration feels rushed, cut words or extend the scene instead of asking the narrator to speak faster.

Choose a voice for the audience

Decide whether the voice should reassure, teach, guide, or persuade. Describe the sound in concrete terms, such as “calm and direct” or “warm, with light energy.” Avoid vague directions like “make it engaging,” and do not choose based only on gender or accent. The right voice depends on the audience, the message, and the performance.

Studies have found different results for human and AI narration. Research on short-video ads found advantages for human narration in that setting. Research in other instructional contexts found that synthetic voices can work well. Test a synthetic voice rather than assuming it will replace a human performance. If you hire a narrator, ask them to record a short passage from the script, including a technical line and a change in tone.

Make a narrated video with explainroo

explainroo uses the open Kokoro voice model to read English scripts in natural American or British English voices. You do not need an account or API key for Kokoro. Narration is available in English and seven other languages. A coding agent creates script.md with the spoken words and scenes.js, the JavaScript code that draws each scene. Whisper, a speech-recognition model, marks when each word is spoken so the visuals appear at the right time.

To try it, paste this prompt into a supported coding agent, replacing the bracketed 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 and checks the video, then gives you an MP4 file. Describe any changes you want, and the agent can make them. explainroo draws scenes in Chrome, adds background music and sound effects, and lowers the music during narration. It checks scene stills and a contact sheet, looks for overlapping or cut-off text, and checks for mispronounced words.

Choose from six built-in visual looks and set the pace in video.json. The available formats are wide, tall, portrait, and square. Tall, portrait, and square videos include word-by-word captions. If the voice mispronounces a product name or technical term, correct the script or pronunciation in your instructions to the agent, then check the next render.

Check the final mix and rights

Listen to the full video, not just a voice sample. Check that the narration matches the visuals, names and numbers are correct, and music or effects do not cover important words. Listen on headphones and play the video on a phone or laptop. Research on listening effort shows that difficult listening conditions make speech harder to understand.

Include accurate captions wherever you publish the video. Captions show spoken content as text, but the narration still needs to be clear. The W3C guidance on prerecorded captions says captions should include dialogue and relevant sound information without covering important visuals.

If you hire a narrator, agree on where you can use the recording and for how long. Also confirm whether the fee covers paid advertising and how script changes or pickup lines will be handled. Usage terms differ for non-broadcast projects and paid placements, as the Voices rate guide describes. The coding agent is a separate service with its own pricing, while explainroo is free and open source.

Free and open source

Let your AI agent make the video

explainroo lets a coding agent like Claude Code or Codex make narrated explainer videos and product demos. The voice, the word timing and the rendering run on your own computer, with no API key and no cost per video. Optional outside services use your own key.

Copy this into your AI agent

Make me a short explainer video about [your topic]. Use explainroo for it: run npx --yes explainroo docs and follow what it prints.

Getting started Example videos GitHub

Made with explainroo: How explainroo makes a video

More on explainer videos

All guides on explainer videos