Church services · Conferences · Seminars

What is said on stage, in a language your audience can read — without waiting.

Not captions that wait for the sentence to end. Live translated captions that stream by meaning, the way a simultaneous interpreter works. Feed it the sermon notes or the slides and names and terms get more accurate. Switch it on ten minutes before, and reuse one link every week.

  • Nothing to install — a browser and a microphone
  • Volunteers run it with a passcode, no account
  • Same captions on OBS, LED walls and phones (QR)
  • One caption language per channel today — Korean, English, Japanese, Chinese or Spanish
LIVE · Sunday 11am · KO → EN 00:00
illustration (simulated) glossary cat-church · Grace Church display source + translation

A simulated illustration of how it behaves — the translation starts flowing before the speaker finishes, then settles when the clause is final.

A late caption is a transcript, not interpretation.

Ordinary auto-captions translate only after the sentence ends, so the audience is always one sentence behind the speaker. Interpreter booths and receivers are accurate, but they cost budget and people every single week.

  • Auto-captions translate after the sentence → 3–6 s behind
  • Interpreter + receivers accurate, but hundreds of dollars per service and gear to manage
  • ISAAC Live clause-level streaming → moves with the speech, billed per minute

How it works

Three steps. A week is enough before your first service.

  1. 1

    Create a channel

    A channel is one stage (one microphone). Sign up with an email, name your hall, and you get the audience link and QR, the operator console link and the OBS source right away. Same links every week.

  2. 2

    Add your material

    Drop in sermon notes, slides or a YouTube link of a past service. We extract names, terms and context and keep them on the channel. The original text is never stored. We walk you through the first one.

  3. 3

    On the day, press start

    The media team opens the console, enters the passcode and starts. Source + translation, translation only, source only and the number of context lines can be changed on the spot. A transcript (SRT/TXT) remains afterwards.

Features

Moves with the speech; terms come from your material.

Streaming by meaning

A provisional translation flows before the clause ends and settles quietly once it is final. The screen never shrinks, so it stays readable.

Channel briefing · 3-tier glossary

Shared (scripture, industry) < category < your channel — the more specific term wins. New terms are learned automatically every week.

The operator chooses the display

Source + translation, translation only or source only; context lines; font size. Change it in the console and every screen follows instantly.

OBS · LED · phones

OBS browser source (transparent lower third), LED side panel, and a mobile page the audience opens from a QR code. No app to install.

Volunteers need no account

Operators enter the console with the channel passcode. Only admins sign in by email to manage channels, members and billing.

Rehearsal studio

Run captions on a recording of a past service, fix wrong terms on the spot, and see the quality before the first live service.

Screens

What the audience sees, what the operator sees.

Use cases

One engine, two stages.

⛪ Church services

Every member follows the same sermon, in their own language. Sunday captions for multicultural members, international students and spouses. Add the sermon notes each week and names, places and church-specific terms come out right.

  • One channel = one hall, the same QR every week
  • A media-team volunteer switches it on ten minutes before
  • Monthly plans; captions never stop mid-service

🎤 Conferences · Seminars

International speakers, multilingual audiences. Add the slides and speaker bios and product names and technical terms come out right. We help with on-site setup and rehearsal.

  • LED side panel, lower third and phones at the same time
  • Prepaid minute packages per event
  • Transcript and SRT after each session

Pricing

Billed by minutes used. Channels are unlimited.

The unit is minutes used. Going over the included minutes never stops the captions — overage is settled next month. Every new organization gets 120 free minutes.

Basic

₩39,000 / month

360 min / month (6 h)

One 1-hour service a week

  • 1 caption language
  • Transcripts kept 30 days
  • Unlimited channels · shared glossaries
Start free

Plus

₩149,000 / month

2,100 min / month (35 h)

Two or more services a week · several stages

  • All 5 caption languages
  • Transcripts kept 1 year
  • Priority support
Start free

Excluding VAT. Overage is ₩100 per minute, settled next month. Onboarding help (audio hookup, glossary, first service) is a one-time ₩200,000–500,000. Several caption languages at once is in preparation and will switch on within each plan's language count. Events (conferences) are quoted separately with time passes and on-site operation. Prices may change before launch.

Sermons and slides are sensitive material.

So it was built this way from the start.

  • Originals are not stored — documents and subtitles you add are reduced to terms and a briefing, then discarded.
  • Transcripts auto-delete — after the plan's retention period (30 days by default); admins can delete them at any time.
  • Never used for training — audio and material go to the OpenAI API only to produce captions and are not used to train models.
  • No sign-in for the audience — the page opened from the QR code asks for no account or personal details.

FAQ

Questions people ask

Do we need internet?

Yes. Speech is transcribed and translated in the cloud. Ordinary venue internet (1 Mbps upload or more) is enough, and the connection re-attaches by itself if it drops.

What equipment do we need?

A laptop (or iPad), Chrome and a microphone. If you have a mixing console, a USB audio interface with a line feed is recommended. Output goes to an OBS browser source or a window on the projector.

Which languages?

Captions between Korean, English, Japanese, Chinese and Spanish. Today a channel has one spoken language and one caption language. Several caption languages at once is on the roadmap.

How accurate is it?

High for ordinary speech; names and specialist terms improve with the material you add, which is why we do the first intake and rehearsal with you. Captions are machine-generated and the audience page says so.

What if we run out of minutes mid-service?

Captions keep going. Overage is billed per minute next month; only when you hit the monthly cap (twice the included minutes) can a new session not be started. The console always shows the remaining minutes.

We run several halls or events.

Create as many channels as you need. Billing counts minutes used, never channels. Agencies manage several event channels under one organization.

Contact

Rather talk first?

Tell us about your church or event and we reply within a day. We do the first material intake and rehearsal with you.

Let your audience read, starting this Sunday.

Five minutes to sign up, one minute for the first channel. Rehearse on 120 free minutes, then decide.