Skip to content
Marketing & campaignsElectronics & SensorsMachinery & Plant EngineeringConveying & IntralogisticsChemicals & Process Industry+2

Subtitles and voice-over in every language

demi translates product, training and recruiting videos with your glossary and creates subtitles and a voice-over for every language.

Today

Product films, training and recruiting videos usually exist in just one language. Every additional language means a translator, a voice studio and editing, often weeks per video, so the original stays the only version. Distributors and customers abroad get videos they do not understand, and training sessions are interpreted on site, with terms that do not match the manual.

An example

Valbrecht has 14 training videos for the VZ 248 infusion pump, all in German. Its distributors in France, Italy, Spain and Poland want them in their language. The agent creates subtitles and voice-overs in four languages, using the terms from the approved instructions for use, such as occlusion alarm, bolus and flow rate. In eleven places the French text runs longer than the scene, and the agent proposes shorter wording; it also marks two scenes because they still show the menu of the old software version. After review by the distributors' native speakers and approval by regulatory affairs, all 56 versions are ready in three weeks; a single language used to take two months.
Fictional company, typical workflow.

Your data, wherever it lives

Your videos sit today in the media library, on drives and in SharePoint, the valid technical terms in approved manuals and data sheets as PDFs in the DMS and in the glossary, often a spreadsheet or a wiki page. demi reads, understands and links them, with your rights. So the AI creates subtitles and voice-overs in days instead of months, with the same terms as the manual, and native speakers check before anything is published.

Where the data lives today

  • Photos, video & audio
  • Folders & drives
  • SharePoint
  • DMS & archive
  • PDFs
  • Wiki & intranet
  • Office files
Data LayerOne view of everything, with your rights

How demi solves it

Three layers in this order: connect the data first, then the rights, then the AI.

  1. 1 · ConnectData Layer

    The case needs data from the media library, drives and project folders, plus approved manuals and data sheets from the DMS and your glossary. demi replaces nothing and links every video to product, software version and the approved translations of technical terms. After approval, demi stores subtitle files and voice-over versions next to the original in the media library.

  2. 2 · PermitPermission Layer

    Internal training videos stay internal, and each distributor sees only the versions for its market. The agent uses a real person's voice only with their written consent. A native speaker checks every language version, and only after approval, for medical devices also by regulatory affairs, does it leave the company.

  3. 3 · ActIntelligence Layer

    The agent transcribes the soundtrack, translates it with your glossary and the terms of the approved manuals, and leaves numbers, units and product names unchanged. It measures each passage against the scene length and proposes shorter wording where a language runs longer. It translates on-screen text and lower thirds too, and marks places where the video no longer matches the current product.

How it runs

  1. 1

    You choose the video, target languages and type: subtitles, voice-over or both.

  2. 2

    The agent transcribes the soundtrack and translates it with your glossary and the terms from approved manuals.

  3. 3

    It fits each passage to the scene length and translates on-screen text and lower thirds.

  4. 4

    Native speakers check original and translation side by side, marked places first.

  5. 5

    After approval, the agent creates subtitle files and voice-over versions and stores them in the media library.

What changes

  • Every video in every language, without a new production

  • The same technical terms as in manual and data sheet

  • International versions in days instead of months

What does it bring you?

Check it with your own numbers

Every starting value is an assumption you can change. The calculation sits right below, step by step.

Assumption, editable

Your numbers

Result

+144 hfreed per year

Assumptions and calculation

demi creates subtitles and voice-over with your glossary; native speakers check and you approve.

5 h → 3 h per case

How does it run today?

What is already digital saves less: the factor lowers the share demi brings.

How we calculate

6 video language versions × 5 h × 40% × 1.0 (manual) × 12 months = 144 h

Value per year, before demi costs

Freed per year+144 h

Assumption: labour cost of €49.50 per hour in German manufacturing in 2025 (Destatis), plus workplace, IT and overheads.

= Value per year, before demi costs

+€9,360

gained per year

144 h ÷ 1,760 h per role = 0.1 full-time roles

An estimate, not a promise. The starting values are assumptions for an industrial company with 150 to 500 people.

Combine with more cases and get the package

Questions about this use case

Does the voice-over sound the same in every language?

Yes. You choose a synthetic voice once, and it speaks every language with the same tone. If the voice of a person from your company is to be used, that requires their written consent. You define the pronunciation of product names and technical terms in the glossary.

What happens when the translation runs longer than the scene?

The agent measures each passage against the scene length. If a translation does not fit, it proposes shorter wording that says the same thing instead of making the voice speak faster. You see these places in a list and decide.

Who checks the translation of the videos?

A native speaker from your company, at the distributor or at your translation agency, depending on how you work today. The agent shows them original, translation and glossary terms side by side and marks where it was unsure. No version goes out without approval.

Are on-screen text and captions in the picture translated too?

Yes. The agent translates lower thirds, on-screen text and credits and sets them in your template. Text that is part of the picture, such as on a display, it marks; you decide whether the scene is reshot or gets a note.

See this case with your data

In the demo we show you the workflow on your own systems.

Your configurator

No use cases yet

Your configurator is still empty

Add a use case from any card or case page and see what it brings per year, with the calculation. Free, no sign-up, no commitment: try it out and change any number.

Browse use cases