Skip to content
Live captioning

Live captions that go in the signal, not beside it

Generated while somebody speaks and out by whichever door you need: over the picture in your switcher, full-screen over HDMI, on any TV in the venue, and on every attendee’s phone.

outputs from one machine
4outputs from one machine
languages in the catalogue
58languages in the catalogue
audio to NDI out
<1.5 saudio to NDI out
apps for the audience
0apps for the audience

What is live captioning?

The text of what is being said, written on screen a few seconds behind the voice. It does two jobs at once: access (somebody hard of hearing can follow the event) and comprehension. A noisy room, an unfamiliar accent or a technical term stop being obstacles for everyone else. Translate the same text and the language problem is solved by the same mechanism.

The problem today

Four outputs, one event

  1. NDI with a real alpha channel

    Genuine transparency in the frame itself, composited over programme as one more input.

  2. Full-screen HDMI with a chroma key

    For when the venue screen does not pass through a switcher. Key colours are adjustable.

  3. Any TV in the venue, through a browser

    Up to two languages on one screen, set from the URL. No video leaves the operator’s machine.

  4. The attendee’s phone

    Scan a QR code and pick a language, up to 12 at once, from a catalogue of 58.

What the operator controls

The text being transcribed, while it goes out

With low-confidence words flagged, so you know how the event is going out while it goes out.

Custom vocabulary, loaded before

Names, brands and acronyms reach the speech recogniser and the do-not-translate list.

Typeface, size, colour and placement

The same settings on the preview, the NDI output and the HDMI one: what you see is what goes out.

An optional profanity filter

It masks the audience feed and the exported transcript.

What happens when something fails

Recommended plan

Accessibility

If you only need captions in the language being spoken, this is the plan: one microphone, transcription only. The moment translation is needed, Event Pro.

See pricing

Frequently asked questions

What is the purpose of live captioning?

So that somebody hard of hearing can follow the event, and so that a noisy room, an unfamiliar accent or a technical term stop being an obstacle for everyone else. In the public sector it is also an accessibility requirement.

Do attendees need to install anything?

No. They scan a QR code and it opens in the browser. No app, no account, no sign-up.

How fast do captions appear?

Measured against the deployed service, speech reaches us as text in 718 ms median and 1,038 ms at the 95th percentile; untranslated captions are written on screen shortly after, word by word. The product target is under 1.5 s untranslated.

Does it support several languages at once?

Yes. One event serves up to 12 reading languages at a time, across phones and screens, chosen from a catalogue of 58.

Is the transcript stored?

Not on the live path. Text is only stored if you ask for the AI minutes or keep the transcript when closing the app, and it is deleted automatically after 30 days.

For your kind of event

Other ways people search for this

Try one hour free

One hour of captioning and translation on one microphone, no expiry, no card. Enough to prove it on your own rig before you pay.