A public meeting the public can actually follow
A meeting is only open in practice if the people it affects can understand it. Translate it live, to the phones residents already carry, without asking anyone to install or sign up for anything.
Session code
Enter this code at go.anylang.app
K7M2QP
Live translation for open meetings — not in place of an entitled interpreter
This is built for the ordinary business of an open institution: a council session, a consultation on a development, a briefing about a new benefit or a school catchment. One person speaks, residents listen in the language they think in, and a transcript exists afterwards. It is not a substitute for a qualified interpreter where somebody is legally entitled to one — a hearing, a formal proceeding, anything with a legal consequence for the person in front of you. Where that is the case, the answer is a professional on that language, and the same broadcast lets you put one there.
- Residents need no account and no app
- Captions in the language being spoken
- A human interpreter on the language that carries risk
- A transcript and, if you switch it on, a recording
Good morning everyone, and thank you for joining today's global session. It's wonderful to see so many of you connecting from around the world.
Putting it in front of residents
- 1
Pick the meeting that already repeats
The monthly council session is the right first one. It recurs, so the per-language audience figures answer the languages question with evidence instead of census data.
- 2
Decide which language needs a person
If one community is directly affected by what is being decided, put a professional interpreter on that language. While people stay with them you pay the connection rather than the translation, and it is the one that would matter if it went wrong.
- 3
Print the address, not the code
A permanent link keeps one address across every session, so it can go on the notice board, the agenda and the door and stay true next month.
- 4
Keep the record you need
The transcript comes free for every language the AI translated. A full audio recording is a separate switch, because it is the one thing here that costs while nobody is listening.
Where institutions use it
One voice and a public that has a right to follow it.
Council and committee meetings
Broadcast as it happens, in the languages your residents actually speak, with a transcript that shows honestly where it has gaps.
Public consultations
A consultation nobody could follow is a consultation that did not happen. Anyone can join from a link on the notice.
Emergency and safety briefings
When the message has to reach everyone, put an interpreter on the language most at risk and let AI carry the rest of the room.
Residents who are deaf or hard of hearing
Captions in the language actually being spoken, with text size and contrast the reader controls. It takes none of the five language slots.
A record that does not pretend
Every language with a listener is kept as text, and the stretches nobody transcribed are marked as breaks rather than run together into a smooth false account.
From the chamber's own desk
Push audio from the encoder that already streams the session. Nothing depends on a laptop staying awake at the back of the room.
At a glance
- Fits
- Open meetings, consultations
- Not a substitute for
- An entitled interpreter
- Resident cost
- Nothing
- Data separation
- Per institution, in the database
Common questions
Does this satisfy a legal requirement to provide an interpreter?
No, and you should not buy it for that. Where a person is legally entitled to an interpreter, the entitlement is to a qualified human being, and an AI stream does not become one because it is convenient. What we can do is put a professional on that specific language on the same broadcast, at no translation cost, while AI covers the rest of the room. Take the legal question to whoever advises you, not to a vendor's marketing page.
Where is the audio processed and what is kept?
Audio is translated as it passes through and delivered live. A recording is kept only if your institution switches that on for itself. The transcript covers the languages that had a listener. Each organization's data is separated at the database level, not only in the application, and a tenant cannot read another's rows even through a bug in the application.
Do residents have to identify themselves?
No, and that is deliberate. A listener opens a link or types a six-character code and picks a language — no account, no email, no app. Requiring people to register in order to follow a public meeting would defeat the point of holding one.
How do we procure this?
You buy credits, and an invoice is issued for every purchase — in lei for Romanian customers and in euro for everyone else, following the country on your billing details. There is no subscription, no per-seat licence and no minimum, so there is no annual commitment to put through a committee before you can test it.
What if the meeting runs long?
A single broadcast runs up to three hours. Credits are spent by the minute per language while people are listening, so a long session with a small audience on one language costs about what you would expect and nothing for the languages nobody chose.
Can two departments broadcast at the same time?
One organization runs one live broadcast at a time. Parallel sessions need a separate organization account each — tell us what your schedule looks like and we will help you set it up before the day rather than on it.
Try it at the next session
30 free credits, no card, and nothing for residents to install.