Lightning
Pay the invoice amount in sats. Scan the QR code with any Lightning wallet.
Transcript Pack by Rowan Vale
One flat price by audio length. Your browser does the audio work; you get three ready-to-use files and a private download link.
Up to 60 minutes per pack. No account needed. Outputs are AI-generated and may contain errors.
Get a pack
Choose an audio file or paste a public audio URL. We decode it in your browser, pick the right tier, then show you the price before any upload happens.
Pay the invoice amount in sats. Scan the QR code with any Lightning wallet.
x402-capable wallets or agents can
POST /api/x402/t10
to receive a one-time token. No browser wallet connection is
needed here.
Paste the token from the x402 response to start transcribing.
DEV ONLY
Payment bypass is available because this page is running in local development.
Keep this page open. Chunks upload two at a time, and the show notes are generated after the last chunk.
Preparing upload…
Pricing
One flat price per pack. Pay in sats over Lightning, or in USDC on Base with an x402-capable wallet or agent.
| Audio length | Lightning | USDC on Base |
|---|---|---|
| Up to 10 min | 210 sats | $0.10 |
| Up to 30 min | 500 sats | $0.25 |
| Up to 60 min | 900 sats | $0.45 |
Maximum audio length is 60 minutes. The exact tier is chosen from the decoded duration before payment.
Free samples
Download sample files from a Transcript Pack run. Outputs are AI-generated and may contain errors.
Privacy
For agents and developers
Agents can buy a one-time job token with x402 (USDC on Base, no account or API key), then upload audio chunks over HTTP.
POST /api/x402/t10 ($0.10,
up to 10 min), /api/x402/t30 ($0.25, up to 30 min) or
/api/x402/t60 ($0.45, up to 60 min) with body
{}. An unpaid call returns 402 with the
payment requirements; after paying you get
{"token":"…"}.
POST /api/job with
{"token","durationSec","chunkCount"} →
{"jobId"}.
POST /api/job/{jobId}/chunk/{i}
with Content-Type: audio/wav, for i = 0…n−1.
POST /api/job/{jobId}/finish, then
GET /api/job/{jobId}/transcript.txt,
/captions.vtt and /shownotes.md.
ffmpeg -i episode.mp3 -ac 1 -ar 16000 -c:a pcm_s16le -f segment -segment_time 59 chunk%03d.wav
curl -X POST /api/job -H 'content-type: application/json' \
-d '{"token":"<token>","durationSec":125.4,"chunkCount":3}'
curl -X POST /api/job/<jobId>/chunk/0 -H 'content-type: audio/wav' --data-binary @chunk000.wav
curl -X POST /api/job/<jobId>/finish
curl /api/job/<jobId>/transcript.txt
Machine-readable: /openapi.json · /llms.txt · /.well-known/x402 · /api/config
FAQ
Audio is decoded in your browser and sent in 1-minute chunks only to the transcription model. Files are kept for later download via a private link.
A transcript (.txt), captions (.vtt) and show notes (.md), plus a private result page with download links and previews.
Pay in sats with Lightning, or use USDC on Base through an x402-capable wallet or agent. The price is shown after your audio is analyzed.
No. Outputs are AI-generated and may contain errors, especially with accents, names, music or poor audio. Review before publishing.
One pack covers up to 60 minutes of audio. Longer files are refused before payment.