Upload once. Everything else happens for you.

What runs when you upload

Step 1 — Transcription. Your audio becomes a full transcript through Gladia, a France-hosted speech-to-text provider with a contractual no-training clause. EU-resident, GDPR-compliant, and accurate enough that you'll edit a sentence rather than rewrite a paragraph.

Step 2 — Translation. The transcript is translated into the languages you've configured for your audience. You set the target languages once; every future upload uses them.

Step 3 — Multi-language subtitles. Subtitle tracks are generated for each language and attached to the video. Viewers pick their language; you don't manage tracks.

Step 4 — Local-currency pricing. If the content is paid (a lesson, a course, a paid newsletter post), prices are set across every market you sell into automatically, adjusted from your reference price — a course priced at €100 for Germany shows up as the right number in kronor for a buyer in Stockholm and in dollars for a buyer in Toronto, without you manually mapping every market.

Step 5 — Catalog entry. Everything joins your reusable content catalog. The transcript and translations are indexed for AI semantic search. The video becomes available for use in any course, any module, any newsletter post. You never re-upload the same content. You never re-transcribe it.

The whole pipeline runs in minutes, not days. You ship faster, you ship in more languages, and you stop paying for four separate tools to do one job.

"Never do anything twice"

This is the framing we use internally and the one we hear back from creators most often. The content catalog is the underlying idea.

Every upload — video, audio, document — becomes a permanent, addressable artefact in your catalog. You can:

• Drop the same lesson into multiple courses without re-uploading.
• Reuse a slide deck across a course module and a paid newsletter issue.
• Surface a community video answer in a course Q&A automatically (via AI semantic search).
• Repurpose a long-form course module into a newsletter issue without re-uploading anything.

The catalog grows with you. After a year of teaching, you've built a searchable, reusable knowledge surface — not a folder of files you forgot you uploaded.

What we don't do with your content

Three commitments, contractually backed:

We do not use your content to train any third-party AI model. Not Mistral, not Gladia, not anyone. The contracts with our providers prohibit it. The relevant clauses are in our published DPA (embedded in our Terms of Use).

The AI doesn't make decisions about your students. Transcription is mechanical. Translation is mechanical. Subtitle generation is mechanical. None of these fall under GDPR Article 22 (automated decision-making affecting individuals). More on the Article 22 stance.

All inference runs in the EU. Mistral, Gladia and the inference path itself are EU-resident. Data does not leave the EU for processing.

None of this requires a leap of faith — it's architectural. Every AI call in the pipeline is a versioned, logged skill, and the documentation falls out of the design. The full account is on our AI-first, compliance-first page.

Pricing — there isn't a separate AI tier

This entire pipeline — transcription, translation, subtitles, regional pricing, catalog — is included in the 5% + €0.50, all-in, alongside everything else it already covers: payments, video hosting, tax, refunds, B2B licensing and AI search.

There is no AI add-on, no transcription credit cap, no translation quota. Upload as much as your business needs. See pricing.

What you still control

Review and edit the transcript before it's published, if you want. Quick keyboard-driven editor.

Approve translations for languages where literal translation isn't enough (technical terms, brand voice).

Override the regional price for any market if PPP adjustment isn't right for your offer.

Exclude videos from the catalog if a particular file shouldn't be searchable or reusable.

Defaults are sensible. Overrides are easy.