Sermon Transcription FAQ

Standard tier is $0.006 per minute of audio (about $0.27 per 45-minute sermon). Premium tier with speaker identification is $0.02 per minute ($0.90 per sermon). The Pro Monthly plan is $29/month with 1,000 Standard + 300 Premium minutes included. First 10 minutes are free — no credit card required.

Yes. Every new account gets 10 minutes of free transcription with no credit card required. After that, pay only for what you use at $0.006 per minute on Standard.

No. There are no setup fees, no monthly minimums, and no contracts on the pay-as-you-go tier. You're charged per minute of audio transcribed.

Pricing is already 250× cheaper than Rev.com. For institutional volume above 5,000 minutes per month, contact hello@sermon-transcription.com for custom pricing.

Yes — 250× cheaper. Rev.com charges $1.50/min for human transcription or $0.25/min for AI. Sermon Transcription's Standard tier is $0.006/min. A 45-minute sermon costs $0.27 vs $67.50 at Rev.com.

Standard tier (OpenAI Whisper) achieves 99% accuracy on clear audio (about 1.8% word error rate). Premium tier (ElevenLabs) achieves 99.5%. Both match or exceed professional human transcribers on typical sermon audio.

Yes. Modern AI transcription handles standard theological terminology (sanctification, propitiation, justification, imputation) and Bible book names (Habakkuk, Ecclesiastes, Zephaniah) accurately. For unusual proper nouns, we use prompt seeding to bias accuracy.

AI transcription tolerates reasonable noise, but the best accuracy comes from a lapel mic close to the speaker. If your audio has heavy noise, music bleed, or multiple overlapping speakers, use the Premium tier for better results.

Yes. Both tiers handle quiet passages, prayers, and pastoral asides as long as they're audibly captured by the recording.

MP3, MP4, WAV, M4A, MOV, AAC, FLAC, OGG, WebM, and most other major audio and video container formats. If a standard player can play it, we can transcribe it.

The underlying APIs limit individual requests to 25 MB. We handle automatic chunking for larger files — a typical 60-minute sermon at standard quality is well under any practical limit.

Every transcription returns plain text (.txt), SubRip Subtitle (.srt), WebVTT (.vtt), and optional verbose JSON with word-level timestamps. Premium tier output includes speaker labels.

Yes, in the Premium tier (powered by ElevenLabs Audio Intelligence). Standard tier does not include diarization natively.

Yes. The verbose JSON response includes timestamps for every individual word, enabling karaoke-style captions, exact-quote linking, and rapid clip extraction.

Standard tier supports 90+ languages including Spanish, Portuguese, Korean, Mandarin, Tagalog, Haitian Creole, French, German, Italian, Russian, Arabic, and more. Premium tier supports 100+ languages with auto-detection.

Yes — Premium tier auto-detects the spoken language. On Standard tier, specifying the language explicitly improves accuracy for non-English sermons.

Transcription captures speech in its original language. For translation, pair the English transcript with DeepL, ChatGPT, or Claude. We're working on integrated multilingual output for a future release.

A 45-minute sermon completes in 3–5 minutes. A 90-minute service typically takes 6–10 minutes. We process audio at roughly 10× real-time speed.

Not yet — the current service is batch-only. Sermon Transcription processes recorded audio. For live captioning, look at YouTube auto-captions or Otter.ai's live stream feature, then re-caption with a proper SRT post-event.

Both. Solo pastors building sermon archives, multisite networks needing standardized output, and seminaries archiving lectures all use the service. Pricing scales linearly with usage.

Absolutely. Sermon podcasters use it for show notes, chapter markers, episode SEO, and accessible transcripts. The cost is the same: about $0.27 per 45-minute episode.

Digitize the recordings to MP3 or WAV, then upload. Archive projects of hundreds of old sermons typically cost under $50 total on Standard tier.

Yes. Audio is processed once and deleted from our servers within 24 hours of completion. Transcripts are retained on your account until you delete them. We never share audio or transcripts with third parties.

No. We operate under OpenAI and ElevenLabs enterprise API terms that explicitly exclude API inputs from model training. Your sermons are not used to train anyone's models.

Yes. Audio is processed in-region where possible, deleted within 24 hours, and never sold. We respond to data deletion requests within 30 days. Full privacy policy at our /privacy page.

You do. The transcript is a derivative of the original audio recording, which is owned by the speaker (pastor) and/or the church. Our terms of service do not claim any ownership of your transcripts.

Legally, the speaker holds copyright on their spoken work. Transcribing for internal church archive purposes is widely accepted under fair use. For public republishing, get the speaker's written permission.

Worship music typically requires CCLI or similar licensing for reproduction. Public-domain hymns can be freely transcribed. For commercial music, transcripts of lyrics may still fall under copyright.

Yes. SRT and VTT output upload directly to YouTube, Vimeo, and HTML5 video players as closed captions, meeting WCAG 2.1 Level AA captioning requirements.

Yes — and this is one of the biggest reasons churches subscribe. Full transcripts serve members who prefer reading; closed captions on video serve those watching with sound off or with hearing loss.

Yes. Most churches paste the transcript directly into their CMS as a blog post. We also expose webhooks for automated CMS publishing once you're on the Pro tier.

Yes — the Sermon Transcription API is the same API powering our web UI. See our API documentation page for endpoints, authentication, and code samples.

Yes. The web UI supports drag-and-drop of multiple files. For large archive projects (hundreds of sermons), the API is the cleaner path.

Otter.ai is built for meetings — speed and live captioning. Sermon Transcription is built for sermons — theological vocabulary, scripture references, batch-quality output, lower per-minute cost. For weekly sermon publishing, Sermon Transcription is 25× cheaper than Otter's effective per-minute price.

Rev.com is general-purpose transcription. Sermon Transcription is church-specific: theological accuracy, sermon-optimized AI prompts, and 250× cheaper than Rev's human tier.

Descript is a transcription-aware video editor — best if you want to edit video and audio by editing text. Sermon Transcription is a pure transcription service: simpler, faster, and dramatically cheaper for archive workflows.