Transcribe audio in a Make scenario

Three built-in modules — HTTP, Sleep, HTTP — and a filter. There is no AudioTranscribr app to install because none is needed: it is one JSON request and one JSON reply.

Your first transcription is free · then 5 minutes a month · no card

The modules

HTTP → Make a request: URL https://api.audiotranscribr.com/v1/transcriptions, method POST, header x-api-key = your key, body type Raw, content type JSON, request content {"url": "{{1.mediaUrl}}"}. Turn on Parse response. The reply has an id.
Tools → Sleep: 60 seconds for recordings up to an hour (a YouTube link is usually done in 5).
HTTP → Make a request: URL https://api.audiotranscribr.com/v1/transcriptions/{{2.data.id}}, method GET, same header, parse response on.
Filter on the next route: status equals completed. For long recordings add a second Sleep + GET pair behind a route whose filter is status equals transcribing.
Use the files: results.txt, results.srt, results.vtt, results.pdf and results.json are plain links that work for an hour without a key. Pass one to HTTP → Get a file, then to Google Drive, Dropbox, email or an AI module.

What the status module returns

json

{
  "id": "tr_…",
  "status": "completed",
  "duration_seconds": 1835.2,
  "language": "en",
  "summary": "…",
  "results": {
    "json": "https://…", "srt": "https://…", "vtt": "https://…",
    "txt": "https://…", "pdf": "https://…"
  }
}

The summary field is a few sentences about the recording, handy for an email subject or a Slack message without downloading anything.

Ideas

Watch a Google Drive folder → transcript PDF back into the same folder.
New row in a Google Sheet with a YouTube link → transcript text in the next column.
New Typeform or Jotform upload → transcript emailed to the respondent.

Pricing

Your first transcription is free, then 5 minutes a month. Paid plans start at $9 a month for 10 hours of audio. All plans →

Transcribe audio in a Make scenario

The modules

What the status module returns

Ideas

Pricing

More guides