Engage
What is live captioning?
Displaying spoken session content as text in real time, either produced by trained captioners or generated by speech recognition.
Also called: closed captions, real-time captions, cart captioning, live captions
Live captioning turns speech into on-screen text during sessions, streams and webinars. Human captioners, often called CART providers, deliver high accuracy for names and jargon. AI captioning uses automatic speech recognition and is cheaper and faster to set up, with accuracy that varies by audio quality, accents and vocabulary.
Captions support deaf and hard-of-hearing attendees, people in noisy rooms and non-native speakers. Many tools also translate captions into other languages and keep a transcript for on-demand content.
When buying, check accuracy on your own recordings, custom vocabulary support, latency, how captions display in-room and in the virtual player, and whether the output helps you meet WCAG requirements.
Why it matters for event marketers: Captions make events more accessible and are increasingly expected in accessibility reviews during procurement.
How it works
Live captioning works by turning session audio into text with only a few seconds of delay. A feed from the room's sound desk or the streaming platform goes to either a human captioner or a speech recognition engine. Human captioners, often called CART writers, use stenography or voice writing; AI engines transcribe automatically, sometimes with a custom vocabulary of speaker names and product terms.
The text is then delivered where attendees need it: on a screen beside the slides, inside the virtual event player as closed captions, or on attendees' own phones through a link or QR code. Many services also translate captions into other languages and keep a transcript that becomes captions for the on-demand recording.
What to look for in software
Test accuracy on a real recording from one of your sessions, including speaker names and jargon, and check whether you can upload a glossary beforehand. Ask about latency, how captions appear in-room and in your streaming or webinar platform, and whether attendees can view them on their own devices.
For high-stakes keynotes, look for the option to switch to human captioners or add human review. Also check languages for caption translation, transcript export formats such as VTT or SRT, and whether the vendor can show how its output supports WCAG requirements.
Examples
3Play Media offers live captioning with tuned speech recognition or professional human captioners, plus translated captions. Verbit's Venue Live delivers real-time captions and transcripts to screens and phones at in-person and hybrid events, with custom vocabularies. AI-Media's LEXI toolkit shows captions on venue screens or attendees' devices, and Wordly provides AI captions and translation that attendees open by QR code.
WCAG 2.2 success criterion 1.2.4 (Captions, Live) is a Level AA requirement for live audio in synchronized media, which is why captions often come up in accessibility reviews.