Skip to content
Government and public sector

Public meetings anyone can follow, from the front row or from a phone

A council meeting is public by definition, and yet residents cannot follow it: because they do not speak the language, because they do not hear well, or because they are watching the stream from home.

The problem, in procurement language

How it fits a council chamber

  1. It takes the PA feed that already exists

    The same signal that already feeds the recording. Nothing in the room changes.

  2. Captions go into the stream

    Over NDI when the meeting is produced with a switcher, or over HDMI to the room screen when it is not.

  3. Residents follow on their own phones

    With a code that can be printed on the agenda, not only projected.

  4. The clerk gets the minutes and the transcript

    With the real time of each intervention and, with several microphones, the name of who spoke.

What the resident gets

Text, audio, or both

Someone hard of hearing reads; someone who does not speak the language listens in theirs.

Fifty-plus languages, no permission needed

No sign-up, no email, nothing to install.

From the chamber or from home

The same link serves the room and the stream.

Reading settings on their own device

Text size and high contrast — what makes access real rather than nominal.

The part that actually saves work

AI minutes as a first draft

Agreements, decisions, agenda items and commitments, each with the time it was said. It does not replace the clerk; it removes the blank page.

Captions for the meeting video

SRT or VTT, ready to upload with the recording.

A record of who spoke

With one microphone per speaker the name is typed by the technician and travels into the minutes. No algorithm guesses it — which is the difference between useful minutes and contestable ones.

What a tender asks and few can answer

Data is processed in the European Union

The voice goes to a European speech endpoint and so does the translation. In a Spanish public tender this is the first criterion, not a technicality.

The live path stores no transcript at all

What is said in the chamber is not kept on any server of ours. The single exception is the AI summary, generated only if somebody asks for it and deleted automatically after thirty days.

The sub-processor list is public

Published on the site, naming each provider and where it processes. A GDPR requirement, met rather than promised.

The price is published

You can budget and compare without a sales call first. We invoice against a purchase order.

Recommended plan

Accessibility

Live transcription on one microphone, which is the shape of a meeting with a lectern and an ordered speaking list. Event Pro when other languages are needed too.

See pricing

Frequently asked questions

Where is the data processed and stored?

Speech recognition goes to the provider’s European endpoint and translation to a European one. The live path persists nothing. If somebody generates AI minutes, that transcript is stored, encrypted, in the EU, and deleted automatically after thirty days.

Does it count as accessibility evidence?

There is a downloadable audience report per session, with devices and languages. On formal conformance with the applicable standard, talk to us and to your advisers: we do not yet issue a formal conformance statement, and we would rather say so.

Can it be bought on an annual budget line?

There is an annual plan. Write to us if you need an hours pool with annual validity to fit a budget year.

Does anything need installing in the chamber?

A desktop application on the machine that already runs the stream, Windows or Mac. Nothing in the room hardware.

Other solutions

Run your next event free

One hour of full-featured captioning and translation, no expiry, no card — enough to prove it on your own rig before you pay.