CHURCHAUDIOAI GUIDE

From signup to your first live service

A step-by-step guide for administrators, speakers, and listeners.

Before you start

Account and email confirmed
Subscription active or trial available
Voice deposit funded
Speaker voice prepared
Listener phone and headphones connected for a test

1. Account and billing

Choose a plan, create an account, confirm your email, and finish onboarding. Pay the subscription by card. The subscription unlocks platform features; a separate deposit pays only for voice minutes actually generated.

Use the plan and deposit buttons in Dashboard. Stripe securely handles checkout. Cards, invoices, cancellation, and payment methods are available under Subscription, card and invoices.

2. Prepare a voice

Open Settings → Voices. Choose a standard voice, record 60–90 seconds, or upload a clean sample. Wait for Ready on the selected voice route and play its test preview.

Record normal speech in a quiet room without music or echo. Your organization’s voices are private.

3. Create a service

Enter the service title, speaker, and source language. Listeners choose their own translation language; the administrator controls the allowed set. Select a voice profile and save.

ChurchAudioAI creates a public listener URL and QR code. Never share the private host link with listeners.

4. Run preflight

Use Desktop Chrome and choose microphone or browser-tab audio. For Google Meet, enable tab audio in the share dialog. Run System Check and verify Audio source, Live connection, and AI providers.

Join from a second phone, choose a language, and test both subtitles and sound.

5. Go live

Display the QR code, ask listeners to wear headphones, and let them select a language. Press Start Translation. Monitor text latency, audio latency, queue depth, and stage health separately.

Pause stops processing temporarily, Resume continues, and Stop ends the service safely.

6. Listener experience

The listener confirms headphones, selects a language, and enables playback. Subtitles arrive before audio. A new listener joins the current live position and never hears the service from the beginning.

If the browser blocks sound, use Sound Check. Subtitles remain available when voice generation is degraded.

7. After the service

End the live session in the host panel. Dashboard provides statistics, transcript, translations, and sermon-content tools. Repeat the service as a template or delete it manually after review.

Saved speaker voices are not deleted with an ended service.

8. Troubleshooting

No subtitles: check input level, source language, and live connection. No audio: check active listeners, voice readiness, deposit, and autoplay permission. High latency: inspect queue, network, and preset.

Allow a few seconds after reconnect. If status stays degraded, end safely and launch a fresh session from Dashboard.

9. Usage billing

Standard voice costs $0.10 per generated minute for each active language. Premium expressive voice adds $0.30. Silence, languages without listeners, and time without generated audio are not charged.

Before TTS, the system reserves a conservative maximum; after generation it charges actual audio duration and releases the remainder.