Frequently asked questions

What does Verstehbar do?#

Verstehbar makes spoken language at events readable in real time: as captions on the big screen and on guests’ smartphones. If needed, it also translates, for example between German and English.

How do live captions and translation work?#

A laptop at the venue receives the sound, for example from the mixing desk. The browser sends it, encrypted, to a specialised speech service (Soniox, processing in the EU). It turns speech into text as people talk and translates it automatically. Our system passes the text on to the big screen and to every phone that has opened the link.

With two languages, the service detects which one is being spoken; both the original and the translation are available. Translation is done by machine, not by people.

How accurate are the captions?#

Usually easy to read, but not free of mistakes. Names, technical terms, strong dialects, unclear speech and people talking over each other are hard. We add names and terms for your event in advance, which improves recognition. Good microphones help most.

Does Verstehbar replace human captioners or sign language?#

No. Human captioners are more accurate, and for many deaf people sign language is their first language. Verstehbar is an addition where there would otherwise be no captions at all.

Do I need to install an app?#

No. Scan the QR code on the screen or poster, or open the link. The captions appear in your browser. You don’t need an account.

Which devices work?#

Current smartphones, tablets and laptops with an up-to-date browser. We have run automated tests with the browser engines of Chrome, Firefox and Safari, including phone sizes. Before the first event we also test on real devices, including Android with Chrome and iPhone with Safari.

For the big screen, any computer with a browser connected to the projector or display will do.

How does access via QR code work?#

Each event has its own link that cannot be guessed. It is shown as a QR code on the screen and on posters in the room. Anyone with the link can read along, without signing up. So the link is not secret, but it can’t be found publicly either. Treat it like an invitation.

Can I change the text size and colours?#

Yes. On your phone: larger or smaller text, light or dark, switch language. Your browser remembers these settings on your device. On the big screen, the event team adjusts font size and contrast to suit the room.

Which languages are supported?#

German, English and French are tested. The speech service recognises and translates 60 languages, including Spanish, Arabic, Portuguese, Turkish, Russian and Ukrainian; the full list is on the home page. If your language is not listed, please get in touch – we will check whether we can support it.

Verstehbar cannot show sign language. How many languages can run at the same time is agreed per event.

What happens to the spoken word?#

Audio is sent encrypted directly from the laptop at the venue to Soniox and turned into text on EU servers. Our own servers never receive audio, only the finished text, which they pass on live to the screen and phones. According to Soniox, content is not used to train its models.

Is audio or a transcript stored?#

There is no audio recording. According to Soniox, live audio is only processed transiently. Our event server keeps a limited caption history in memory and does not archive it. The event team can optionally enable a local transcript in the operator’s browser.

The local transcript captures text from the time it is enabled, including during a live broadcast, and can be exported. It remains in browser storage until manually deleted. The organiser determines its purpose, use and retention and informs affected people beforehand. The privacy policy explains technical operational data and stored context information.

Am I tracked on the website or while reading along?#

No. No cookies, no analytics, no advertising trackers. The caption view only stores your settings (language, font size, light or dark) in your browser.

Read the privacy policy

What do we need on site?#

A laptop with an up-to-date Chrome, a sound feed (ideally a separate output from the mixing desk, for example via a USB audio interface), a stable internet connection and, if you like, a projector or display for the captions. We work out together what fits your setup.

Does it work for hybrid events?#

If the sound of people joining online runs through the room’s sound system, it can be captioned too. We check this in advance in a technical test.

What does it cost?#

During the pilot phase, terms are discussed individually. There is no price list yet.

What does the current pilot status mean?#

Verstehbar is under development. The technology is built and covered by automated tests; its first use at a public event is in October 2026. Organisers can’t sign up on their own yet and there is no price list. We support every event personally and improve Verstehbar after each one.