Event Translate turns live speech into translated audio and live transcripts in 35 output languages. Learn how to connect a microphone or venue mixer and share a QR code with listeners.
Create an account, connect your audio, and click “Start Translation”.
Sign up with email or Google. Event Translate creates your event, unique event link, and QR code automatically.
On the Stage configuration page, select your mixer or audio interface, choose input and output languages, then check the meter while someone speaks.
Start once the meter is moving steadily. The listener page will begin receiving translated audio and transcripts.
Complete these before the event to match the listener page and terminology to your organisation.
Both On-demand and subscription customers can choose a name in Console → Settings → Custom event link, then download the QR code or copy the link. Reuse it across dates and locations while the name and room stay unchanged. Personalised paths and branded subdomains have different plan benefits; see Pricing for subdomain access and permanent ownership.
Boost speech translation accuracy by giving AI the context of your conference. In Settings, describe your event theme and add key company or speaker names with custom translations. Enter them once before the session starts to keep every language consistent.
Reach global attendees simultaneously with real-time translation in up to 35 languages. Choose your target languages on Stage before starting the live broadcast. Each channel delivers a high-quality AI voice (TTS) that dynamically adjusts its speed to match the speaker's pace.
Deliver a cohesive event experience by matching the listener page to your organisation. Upload your logo, set a custom event title, and customize banner theme colours. Attendees see your identity from the moment they scan the QR code.
Listeners join from their own phone browser. No app installation or account is required.
Event Translate prioritises audio streaming. All 35 output languages provide high-quality audio and a live transcript together.
Download the QR code or copy the custom event link from the Console, then place it on presentation slides, event materials, or attendee messages
Listeners scan with their phone camera or open the custom event link to go straight to the event listener page.
Select from the languages enabled by the operator for this event.
Tap play for high-quality translated audio and follow the live transcript at the same time. AI Live Summary is now available too (Beta).
Event Translate works in a modern laptop browser. The key is sending a clean, speech-focused signal from the venue into that browser.
The closer the input is to a clean, dry voice signal, with little room reverb or music, the more clearly Event Translate can recognise and translate it.
For most events, the front-of-house or main mix is a practical starting point. Make sure speech is not buried by music or playback.
Build a separate output that keeps presenter microphones prominent and removes unnecessary instruments, background music, and audience microphones.
When there is no mixer, connect a lapel microphone to the laptop. This works for touring speakers or small venues.
For a test, place a phone or tablet near the lectern. Do not place it in front of a PA speaker or at the back of the room.
Send an AUX, matrix, or Control Room line output into the interface, then connect the interface to the laptop by USB.
If the mixer sends the required mix through CARD OUT or compatible USB audio, connect it directly and select that input in the browser.
Use only a true line input with the correct cable. Most modern MacBooks need a USB audio interface. Never connect a speaker-level output.
A one-page wiring summary, available in English, Simplified Chinese and Traditional Chinese, and switched automatically with the page language. Use it for quick reference on site; the written steps here and in the technical guide remain the complete, screen-reader-accessible instructions.
Download the quick reference image
For a venue, connect the mixer’s speech feed to a laptop through a compatible USB audio interface. For a mobile tour, a phone can capture speech, and a compatible lavalier microphone is recommended for the event. Both operator and listeners need reliable internet, and listeners need headphones for audio. Before using AirPods or another Bluetooth microphone, test the input, sound quality and connection on the actual phone and browser.
Paid plans have no per-session language cap: pick as many of the 35 target languages as the event needs, and each runs its own translation in parallel. Free trial sessions are limited to two. Every extra language consumes credits at its own rate, so a three-language hour costs three times a one-language hour.
Before you press “Start Translation”. Pick a preset theme, then add each term once to the glossary: company names, speaker names, department or product terms, and words that are easy to mishear, with an optional preferred translation per target language. These settings lock once the session is live, so a term added mid-event only takes effect after you stop and restart the broadcast.
Send a clean, speech-focused feed with music, effects and audience mics kept out wherever possible. A post-fader aux or matrix output is ideal because it follows the speaker without carrying the room. Connect it to the laptop through a USB audio interface, or use the mixer’s own USB audio output, then confirm the input meter moves with speech before going live.
No, the number of microphones does not change the bill. If the room’s microphones are already mixed into one feed, use the standard mode. Choose Multi-mic 2× only when two speakers arrive on genuinely separate left and right channels and you want them handled separately. That costs twice the standard rate. Check that the two channels are truly independent before going live to keep speakers distinct and reduce bleed.
Audio tuning yes, translation settings no. Noise gate threshold and voice-detection sensitivity take effect instantly while you broadcast. Source language, speech recognition provider, target languages, event context and glossary all restart the translation when changed, so they stay locked until you stop. Settle them during setup rather than mid-event.
The session adjusts streaming quality and reconnects automatically: after a brief drop, the operator and listeners recover on their own and audio and transcripts resume without anyone re-scanning the QR code. The operator connection carries the source upload, so use wired ethernet, a dedicated SSID or tested portable Wi-Fi; each listener still needs reliable Wi-Fi or mobile data. Do not size the event from an empty-room speed test alone. Rehearse with the expected device count.
Listeners are never billed and never need seats assigned. For planning, allow for 500–1,000 simultaneous listeners per room and roughly 100 kbps of downstream bandwidth per listener; this is a planning guideline rather than a universal capacity guarantee. Actual capacity depends on devices, network conditions, languages and event configuration. Tell us the date in advance if you expect more or need several rooms at once.
Use the free trial to rehearse with your actual microphone, languages and listener phones. For setup help or event-day support arrangements, email contact@eventranslate.com in advance with the date, room count, expected connected listeners and audio equipment. We can then confirm the available assistance and arrangements.
No. Set the source language to Multi-language and the system detects 70+ languages automatically, following a speaker who changes language mid-event or even mid-sentence, including mixed Chinese and English. If the whole event uses one language, you can also select that language directly.
