For churches, schools and events
How to choose AI interpreting for your church, school or event
Live AI interpreting turns the speaker’s words into captions, and often speech, in each listener’s own language. This guide lists what to check before you choose.
Updated 14 November 2026
Thirteen criteria for your gathering
| Criterion | Why it matters | Ask any provider | Espero |
|---|---|---|---|
| Joining | A clear way in helps guests start following the gathering. | How do guests open captions, and what do they need to sign in? | A QR code or viewing link opens captions in the browser; reading needs no app or account. |
| Languages | Match the broadcast’s languages to the people taking part. | Which caption languages are available, and how many fit in one broadcast? | Choose from a 106-language catalog, with up to 10 languages per broadcast, the spoken language included. North Sami is also available as a caption language. |
| Listening | Some guests want to hear the translation while following the event. | Which languages offer spoken translation, and what does listening require? | Listen through your own headphones with an AI-generated voice in supported languages. No account is needed. |
| Several speakers and interpreters | A gathering can include people speaking different languages. | How do we set up several spoken languages and a human interpreter? | Multiple speakers can speak different languages. Add the interpreter’s language and turn on “An interpreter repeats the speech”. |
| Sound | Your venue’s sound source shapes the practical setup. | Which audio inputs can we use at our venue? | Use the device microphone, a mixer over USB or Dante as the audio source. |
| Screen and stream | Many gatherings show captions in the room and stream to people following from home. | Can captions appear on the venue’s screen and on our stream, and can stream viewers follow in their own language? | The projector view shows large captions on the venue’s screen. Stream caption overlay adds them to OBS, vMix or ProPresenter over a transparent background. The link for stream viewers waits for the YouTube or Facebook stream’s delay. |
| Cost model | The billing unit and microphone time help you plan a budget. | What is the billing unit, and how are silence, pauses and VAT handled? | Captions in the spoken language are included. Each translation language and each listening language costs €2.50/h with an annual subscription, €3.00/h monthly or €3.30/h one-off, plus VAT. Captions in the spoken language alone cost one language. Time is used while the microphone is on, but after two minutes without speech the meter pauses and resumes when speech starts. A paused microphone uses no time. If your balance runs out during a broadcast and you have bought hours before, the broadcast continues for up to one hour, and the extra time is deducted from your next purchase. |
| Trying it | A rehearsal in your own setting lets you explore the workflow. | What does the trial include, and can we use it across several sessions? | Try 7 hours free, with no card needed and time usable in parts. A broadcast using only trial hours holds up to three languages, the spoken language included, with AI voice in one of them. |
| Text afterwards | An editable record supports sharing and reading after the gathering. | Which export formats, correction tools and retention periods are available? | Export TXT, HTML or PDF. Corrections update every language, and the text is kept for 30 days from the start of the broadcast. |
| Conversation | Questions help listeners take part in the shared gathering. | How can listeners ask questions and take part around the event? | Every event has its own group before, during and after, with questions in each person’s language. The organizer can show a question on the projector without the sender’s name. |
| Preparation | A shared checklist gives organizers a repeatable routine. | How can we prepare vocabulary, share joining instructions and switch devices? | Use the Before the broadcast checklist, a printable QR poster, an event glossary of up to 30 terms, and songs and special words. Switch devices with Resume broadcast. |
| Data | Know where content is stored and how AI processing works. | Where are records stored, and what are the AI services’ training and retention terms? | Messages are encrypted in transit and at rest, and the primary database is in the EU (Ireland). Espero uses the paid APIs of AI services whose terms state that the providers do not use the content to train their own models; they may keep requests for a limited period for abuse monitoring. Security and privacy |
| Platforms and help | Organizers need a setup they can use and guidance they can follow. | Which platforms and organizer guidance are available? | Espero works on iOS, Android and the web. Follow the organizer guide or request a 15-minute introduction. Organizer guideAsk for an introduction |
Questions to ask any provider
- Can listeners read without installing anything?
- Which languages can one broadcast carry at once?
- What exactly counts as billable time?
- Can we connect our own sound desk?
- Can we correct and keep the text afterwards?
- Where is the data processed and stored?
- Can we try it at a real gathering first?
- How do listeners ask questions?
- Can viewers of our stream follow in their own language?
Test it at your next gathering
- Use your own sound source: connect the microphone or audio feed you plan to use at the gathering.
- Invite a few listeners who read different languages and ask them to follow the same talk.
- Try listening with headphones and check the experience from a listener’s seat.
- Read and correct the text afterwards, then export it in the format your team will use.
How Espero fits
Espero is a multilingual messaging app with live event captions. The speaker talks as usual, and everyone follows in their own language. Messages and written-text translations are free.
AI interpreting: practical answers
What is AI interpreting?
AI interpreting converts live speech into translated captions or speech so people can follow in their chosen language. Espero brings this to event captions, with AI-generated voice listening in supported languages.
Do listeners need an app?
For Espero captions, guests open an event QR code or viewing link in their browser without installing an app or creating an account. Reading and listening need no account. An account is needed for the group conversation; an email address is enough.
How is AI interpreting usually priced?
Budget for each translation language and each listening language: €2.50/h annually, €3.00/h monthly or €3.30/h one-off, plus VAT. The speaker’s own captions are included alongside a translation; using only those captions costs one language. The microphone uses time until two minutes without speech pause the meter. Speech starts it again, and pausing the microphone uses no time.
Can AI interpreting work with a human interpreter?
Yes, Espero supports several speakers in different languages alongside interpreters. Select the interpreter’s language and enable “An interpreter repeats the speech” so the broadcast includes the interpreted speech.
What equipment do we need?
Start an Espero broadcast from a phone or computer with a microphone, or feed sound from a mixer over USB or Dante. Guests read on their own phone or computer and use their own headphones for listening.
How can we test it before we decide?
Espero gives organizers a seven-hour trial without a payment card, and the time can be split between sessions. Rehearse with your venue’s sound source, your chosen languages and listeners who can explore the captions and listening.
Can we use the event text later?
In Espero, the organizer can correct broadcast text and download it as TXT, HTML or PDF during the 30 days after the broadcast starts. An edit updates all languages, keeping the exported record ready to share.