Share the signal
Share your screen or a browser tab with audio right in the browser – nothing to install. On site, a microphone is enough.
Live AI simultaneous translation of professional quality – from Switzerland
Interpreting in the cloud. Live in French, Italian and English.
Share your screen, your Teams or Zoom meeting, or your livestream. Our AI interpreter listens, understands the context and speaks the translation in a natural voice – with the original still audible underneath, like a TV interview.
Just a link for listeners
No app, no account
With subtitles
4target languages: FR · IT · EN · DE
5source languages incl. Swiss German
~15 swith picture, in sync
~2 sLive-Highspeed on site, audio only
How it works
Multilingual in three steps
No booths, no headsets to hand out, no equipment for the audience. You send a signal, we deliver the interpreted versions.
The AI interprets with context
It understands what is being said, knows your brief, your glossary and everything said so far – and renders it sentence by sentence into every target language.
The audience listens
Via a link in the browser: the picture, a natural interpreter voice in the chosen language, and the original quietly underneath. Subtitles optional.
Quality
Works like a real interpreter
Any machine can translate word for word. Good interpreting means understanding, waiting for the right moment, choosing the right terms and speaking naturally. That is exactly what Dolmetscher is built for.
Knows the context
Brief, glossary, a running summary and the latest contributions go into every translation. Names, roles and terms stay consistent throughout the event.
Waits for the verb
Like a professional, it briefly holds back half-finished sentences until the meaning is clear – for example when a German verb only comes at the very end. No false starts.
Names and terminology right
Your glossary is binding. Numbers, dates and amounts are rendered exactly, misheard words are recovered from context, and fillers like “um” are dropped.
Understands Swiss German
Dialect, Standard German, French, Italian or English – even mixed. Swiss terms such as “Traktandum” or “Gemeinderat” go into the glossary with your preferred translation.
Natural voice, original underneath
Calm voices in the style of conference interpreters, selectable per language. The original continues quietly in the background and comes back in every pause. If a speaker already uses the target language, listeners hear them directly.
In sync and consistent
As in simultaneous interpreting, the voice starts just after the speaker, in step with the picture. All languages are produced in one pass, so they all say the same thing.
Use cases
Wherever people listen in several languages
For quadrilingual Switzerland and international teams: your audience follows in their own language, without you changing your programme.
Municipal assemblies & parliaments
Debates and votes live for every language group – including the livestream at home.
Town halls & staff briefings
Management speaks once, every site understands in German, French, Italian or English.
Teams & Zoom meetings
Send the meeting window and audio; participants and guests hear the translation via a link.
Events & conferences
Keynotes, panels and award ceremonies: with Live-Highspeed the audience in the room listens on their own smartphones – about 2 seconds behind.
Webinars & presentations
Slides and voice stay in sync – your presentation reaches a multilingual audience.
Schools & parents' evenings
Information for parents in their own language, without extra staff.
Clubs & associations
General meetings and delegates' assemblies across language borders.
Livestreams
Livestreams translated into every language – via a link in the browser.
Features
Everything live translation needs
Professional technology behind the scenes, simple operation up front.
Send from the browser
Share a tab or screen with audio – in Chrome or Edge, nothing to install.
Live-Highspeed on site
Audio only, about 2 s delay: listeners scan a QR code and listen on their own smartphones.
Recordings on request
Every language as a file to download – deleted automatically after 48 h.
Subtitles
On request the player shows the translation as subtitles, timed to the voice.
Chat per meeting room
Questions and feedback from the audience right next to the picture.
One private link per meeting room
Unguessable links: only people with the link can watch. Sending requires a login or a secret send link.
Starts automatically
Translation runs as soon as you share and ends when you stop – you only pay for what gets translated.
Voices of your choice
A dedicated voice per language in the calm style of conference interpreters – with a preview before you start.
On any device
Smartphone, tablet or computer, with fullscreen and language choice in the player.
Swiss operator. Your data under control.
Dolmetscher runs on a self-managed server, operated from Switzerland – on request on your own server, too. The site works without third-party trackers, ads or external fonts.
- Recordings only on request, deleted after 48 h
- AI only while translation is running
- One private link per meeting room
- No cookies, no third-party trackers
FAQ
Frequently asked questions
How long is the delay?
Picture and sound reach the audience with a fixed delay of about 15 seconds. The interpreter uses this time to understand sentences, choose the best translation and voice it in step with the picture – like a TV station broadcasting on a slight delay.
For live meetings with questions, use the chat or plan the delay into the moderation. For on-site events there is Live-Highspeed: audio only, about 2 seconds of delay.
Which languages are supported?
Speakers may use German, Swiss German, French, Italian or English – even mixed, with automatic detection. Interpretation is into French, Italian, English and German, in any combination.
What do listeners need?
Just the link and an up-to-date browser on a smartphone, tablet or computer. No app, no account, no sign-up. They choose their language in the player and can turn on subtitles.
How good is the translation?
Dolmetscher uses current, leading AI models for speech recognition, translation and speech synthesis, and works with context like a professional. The biggest difference comes from a short brief (occasion, names, roles) and a glossary of your terminology.
Like any machine it can make mistakes, especially with poor audio or people talking over each other. For legally binding settings such as court hearings, we still recommend human interpreters.
What happens to our data?
Picture and sound are buffered for the broadcast delay. If recording is switched on, we keep every language and the original for 48 hours to download; after that the files are deleted automatically. While translation is running, speech segments are sent to the AI models for processing (speech recognition, translation, speech synthesis). When translation is stopped, no audio leaves the server.
Event settings are stored too, as is the chat until you delete it. The player sets no cookies and embeds no third-party services. For statistics it only counts, anonymously, how many people listen, for how long and in which language.
How do I get Teams or Zoom in?
In the browser (Chrome/Edge), share the whole screen with “Share system audio” (Windows) or the browser tab with “Share tab audio”.
How much does Dolmetscher cost?
You get 30 free minutes to try it out, no credit card needed. After that, pick a monthly plan – from CHF 120.– for 150 minutes – or pay CHF 1.– per minute without a subscription. Only speaking time while translating counts, per target language. Prices excl. VAT. All prices
Is silence billed?
No. Minutes only count while speech or sound is detected with translation on. Short pauses between words and sentences (up to about 1.5 seconds) count; longer silence, breaks and waiting time are free.
Speaking time counts per target language: 10 minutes of speech in French and English make 20 minutes. The statistics show the transmission duration and the billed speaking time for each session.
Which microphone do I need?
The closer the microphone is to the person speaking, the better: a headset, a lavalier or a microphone right in front of the speaker. If there is a PA system, take the sound straight from the mixing desk – not with a microphone in front of a room loudspeaker.
Reverb, echo and noise can cause recognition errors; in the worst case a “not” gets lost. In online meetings, participants wearing a headset are understood much better.
Ready for your multilingual event?
We'll show you Dolmetscher live with your own content – in your language and the languages of your audience.