Privacy-first AI infrastructure.

← Use cases

Analyse the material. Protect the people in it.

Code and compare transcripts with participants masked consistently across the whole folder, and the mapping never leaving your machine.

Start free in the browserDocuments and folders

What you have

P07, Yolanda Martinez Jarque, teacher at Institut Balmes, Barcelona, interviewed 12/05/2026.

What the assistant receives

P07, ⟦PERSON_1⟧, teacher at ⟦ORG_1⟧, ⟦LOCATION_1⟧, interviewed ⟦DATE_1⟧.

Synthetic. The reply comes back with the tokens in it and the real values are restored on your machine.

Consistent across files
The same participant keeps the same stand-in through a folder of transcripts, so coding and thematic analysis still track who said what.
Pseudonymisation, stated plainly
Identifiers are replaced and a local mapping can restore them. Whether that meets your board's anonymisation standard is their call; what is factual is that the mapping never leaves the researcher's machine.
Testable for an ethics application
The extension's round trip runs with no network access at all, which is worth a sentence in a protocol precisely because a reviewer can test it rather than take it on trust.

Qualitative material is personal by nature. The value is in what people said, and the risk is that they can be identified as having said it.

Ethics approval usually assumes the data stays inside the research team, which pasting a transcript into a chatbot quietly breaks.

What detection catches, and what still slips through, is measured and published on the benchmarks page.

What people ask first

Is masked data anonymised for the purposes of an ethics application?

Masking as Velum performs it is pseudonymisation: identifiers are replaced and a local mapping can restore them. Whether that meets your board's anonymisation standard is their call, but the operational facts are clear to state: identifying details are replaced on the researcher's machine before any text reaches an AI service, and the mapping never leaves that machine.

Will the same participant get the same mask across many files?

Within a masking session the same detected entity keeps the same stand-in, which is what keeps multi-document analysis coherent. The desktop app is the right surface for folders of transcripts for exactly this reason.

Does it handle non-English transcripts?

Detection ships with country packs for Spain, the US, UK, France, Germany, Turkey and Italy, and measured per-language results are on the benchmarks page. If your material is in another language, the regex floor still catches format-based identifiers, but read the benchmarks before relying on it.
Request a demo

See it work on your own data.

Tell us what you work with and we will walk you through it on a call, on your own files. Nothing leaves your machine while we do.

Your details mask themselves as you go. That is Velum, running in this page. Press the eye to unmask. We still receive them in full.

Your address is used to reply to you, and for nothing else.