City Tours Museums Business Churches How it works Pricing FAQ Guide
Get started free
← Back to homepage

Technology & Privacy

Here we explain in plain terms how Live Interpreter works technically and what happens to your audio data during an event.

How translation works

  1. Create a room

    The host creates a room: a digital space with its own address that listeners can join. A room captures one speaker input and distributes the translation to all listeners (1:N). Hosts can set up as many rooms as they like, for example for different events or venues.

  2. Share a link or QR code

    The host shares the room link or the matching QR code with the audience that wants to receive the translation.

  3. Start a session

    The host enters the room and starts the session. The large button in the room lets you start, pause and resume the translation at any time, for example when there's no need for translation during a workshop or during worship time in a service.

  4. What happens in the background?

    As soon as translation is started, Live Interpreter processes the spoken word in real time. First the speech is recognised and converted into text. The content is then translated into the chosen target languages. After that, an audio output is generated from each translation and played to the respective listeners.

    This creates a continuous live flow:

    Speaker → speech recognition → translation → voice output → listener

  5. Listeners follow live

    Listeners open the link in the browser, choose their preferred language and hear the translation in real time, with only a few seconds' delay. No download, no account, no app.

  6. End the translation

    The host ends the broadcast via the same button. The session can be resumed later at any time.

  7. Leave the room

    When the host leaves the room, all listeners are automatically removed from the room and all transmitted content is discarded. Permanent rooms are kept. The host can re-enter them at any time and start a new session, without having to create a new room for every event (such as every Sunday service).

What we store and what we don't

No permanent recording of your event

Live Interpreter is not a recording or archiving system. Audio is processed for ongoing translation. We do not store any audio recordings or transcripts of your event.

No use of your content for our own training

Your spoken content is not used to train our own AI models. Our aim is to configure the third-party services we use so that your content is not used for model training.

Only technical usage data

So that we can operate, bill for and improve the service, we store technical metadata, namely room metadata (languages, time of creation) and usage metadata such as the number of audio minutes used for billing, the time and duration of a session, or technical error notices. This data helps us run the service reliably and bill usage correctly.

No account for listeners

Listeners need no account of their own, no app and no email address. They join a room anonymously via the link or QR code the organiser provides.

Who can listen?

A Live Interpreter room is not publicly discoverable. Listeners can only enter the room via the direct link or QR code that the organiser shares.

The link alone, however, is not enough to listen in on a translation. A room only transmits audio once the host actively opens the room and starts the live translation. As long as the room is not open or no translation is running, listeners merely see that no broadcast is currently active.

This means: a forwarded link does not allow secret access to live content outside an actively started session. The organiser keeps control over when a room is opened, when translation begins and when the broadcast ends.

Even so, the link should only be passed to people who are meant to attend the event. For new events, rooms can be closed or new links used.

Audio setup: from microphone to translation

In the browser, Live Interpreter accesses the audio input device that your operating system provides. That means: any microphone that appears as an input device in Windows, macOS or iOS can be used. In the host interface you select the desired device from a picker.

USB microphones and headsets

Plug in the device; it appears in the sound settings in Windows/macOS, and you can select it in the browser picker.

Mixing console with USB output

If your church hall or event space already uses a digital mixing console with a USB audio interface, you can take a feed directly at the laptop. Live Interpreter then gets exactly the signal that also reaches the speakers, without a separate microphone in the room.

Dante audio networks

For rooms with a Dante infrastructure we recommend the Dante Virtual Soundcard (DVS) from Audinate or a Dante USB interface such as the Audinate AVIO USB. Both provide a Dante channel as a virtual audio input in the operating system. In the browser picker it appears like a normal USB device.

Microphone permission in the browser

On first use the browser asks for microphone access. Choose „Immer erlauben“ (Always allow). This keeps the permission even after browser updates.

Better recognition of technical terms

Every event environment has its own specific terms: proper names, product names, technical expressions, brand terms or abbreviations. These can be stored before a session, either once in the account profile or specifically per room.

The speech recognition takes the stored terms into account and delivers noticeably more precise translations. Particularly useful for industry-specific talks, technical training or specialist events with their own terminology.

Common questions

Ready to try it yourself?