Localizing explainer and classroom videos: terms, reading level and captions

What makes explainer and classroom videos different

Educational videos in this sense are short, self-contained explainers: how photosynthesis works, why fractions need a common denominator, what caused a historical event. They are watched by children and teenagers, by adult learners and by teachers preparing lessons. Unlike a full online course, which is a product with modules and a platform (see online course translation), each explainer has to stand on its own and is often shared, embedded and reused in classrooms you never see.

That creates three priorities: the terms must match what learners are taught, the language must match their level, and the video must work for learners who read along.

Keep the curriculum's terminology

Every subject has official or customary terms in each language, and they are not always the obvious translation. Mathematics vocabulary and notation differ between countries: the decimal separator, how long division is laid out, and the fact that the German Billion and the French billion mean a million million, not a thousand million. Science terms may be Latin- or Greek-derived in one language and everyday native words in another. A learner who hears a different word from the one in their textbook may not realize it means the same thing.

Build a short glossary with a teacher of that subject in the target country before translating a series, and check every translated transcript against it. Terminology errors are the ones learners are least able to spot for themselves.

Match the reading level and pace

Translations can drift toward formal, adult vocabulary. For a video aimed at ten-year-olds, a reviewer should simplify words and shorten sentences in the translated subtitles where the translation went too formal. Pace matters as well: translated narration in some languages takes longer to say, and if the voice speeds up to fit, younger learners may struggle to keep up. Natural pauses in the source narration give the translated voice room to breathe.

Swap examples learners recognize

Explainers rely on everyday examples: money, sports, food, measurements, holidays and place names. A word problem priced in dollars, a distance in miles or a reference to Thanksgiving works in one country and distracts in another. Spoken examples come through the voice track as translated, though the subtitles can be adjusted. For a series you plan to localize repeatedly, prefer neutral examples in the source script: metric units, invented or no currency, and generic place names.

Captions that help learners

In education, captions do more than translate. Learners who are still developing reading skills or studying in a second language benefit from hearing and reading at once, and captions are essential for deaf and hard-of-hearing students. Consider:

• Same-language captions for the source version, and translated subtitles for each localized version.
• VTT files for web players and many learning platforms, SRT for video platforms and offline players; check what your platform accepts.
• Short lines and a comfortable reading speed for younger viewers, splitting long cues where needed.
• Bilingual viewing: a learner of the source language may watch the original audio with subtitles in their home language, or the dubbed version with source-language captions.

The accessibility side is covered in captions for educational video accessibility.

Example: a science explainer in Spanish and Vietnamese

A review workflow for teachers and editors

1. Agree the glossary of key terms with a subject teacher in the target language.
2. Produce the translated transcript and subtitles, plus the voice version if you need one.
3. Have the reviewer read the translated transcript against the glossary and the intended age group.
4. Correct the subtitle files directly, and note any spoken lines that are seriously wrong.
5. Watch the localized video once at normal speed with subtitles on, checking that diagrams and narration still line up.
6. Record the corrections, so the next video in the series starts from a better glossary.

If a spoken line is seriously wrong, adjust the source narration or rely on subtitles for that video, because editing a subtitle file does not change the voice track.

Where AI localization falls short in education

• Diagrams, labels and handwritten notes on screen are not translated, and in explainers they often carry the key information.
• Spoken math is fragile: formulas, variables and units read aloud can be translated oddly, so check every equation.
• Some subjects are country-specific (history, civics, national curricula) and need adaptation, not translation.
• One synthetic voice for every character in an animated explainer flattens the dialogue.
• For very young learners, a human narrator who adapts tone may simply work better.

Using mydubly for educational videos

mydubly takes an explainer exported from your editor (MP4, MOV, WebM, MKV or M4V, up to 2 hours) and returns a translated MP4 with an AI voice in one of 21 languages, the translated audio, SRT and VTT subtitles, and transcripts in both languages for teacher review. Choose one of the eight stock voices and keep it across a series so learners hear a consistent narrator. If you only need captions, transcript mode produces timestamped transcripts and subtitles at 1 credit per minute. Classroom recordings stay on your device; only the audio is processed. The video localization page explains the overall workflow.

Next step

Choose one explainer you already use often, draft a glossary with a colleague who teaches in the target language, and produce a localized version for one group of learners. Watch how they use it, then extend the series. Start from the video localization page; for teachers with several home languages in one class, videos for multilingual classrooms goes further.