Automate transcription with Make
Nine modules for your scenarios: transcribe a public video link, a voice note or a file from an earlier module, and get the text, subtitles or AI analysis back — with an instant webhook trigger so a finished transcript reaches your scenario the moment it's ready. Runs on the same account and the same REST API as everything else.
Status: built, not yet public
The app is built and has been tested end-to-end against production, but it is not yet in Make's public app directory— adding it from a fresh Make account isn't possible today. The same transcription is available now through the n8n community node or directly against the REST API.
The nine modules
- InstantWatch Transcript Events — a dedicated webhook that fires the moment a transcript finishes — no waiting inside a module.
- TriggerWatch Completed Transcripts — polls for finished transcripts, newest completion first — the reliable fallback where webhooks don't fit.
- ActionTranscribe a Media URL — starts transcribing a public video link.
- ActionTranscribe a File — transcribes a file from an earlier module — a WhatsApp voice note, a Drive file, an email attachment.
- ActionGet a Transcript — status, text, timestamped segments and the AI analysis.
- ActionDownload a Transcript — SRT, VTT, TXT, CSV, Markdown, PDF or DOCX.
- ActionDelete a Transcript — removes it and its stored media.
- SearchList Transcripts — newest first, with a status filter and pagination.
- UniversalMake an API Call — any endpoint the REST API exposes, for anything the modules above don't cover.
Built for Make's 40-second limit
Make stops any module after 40 seconds; transcription takes minutes, roughly half a recording's length for long files. Rather than a module that fails on anything but the shortest clips, the instant webhook trigger delivers a finished transcript straight into a scenario with no module waiting at all — the honest, reliable way to bridge a 40-second limit and a multi-minute job.
Limits & behavior
- Public links only — no private or login-required videos, no live streams. Up to 150 minutes per recording.
- Same limits as the app and REST API: 3 new transcripts per hour, one processing at a time during the beta.
- Authenticates with the same API key as the REST API and the n8n node — one key, one account, one set of limits everywhere.
FAQ
Is the Make integration live?
It's built and has been tested end-to-end against production, but it isn't in Make's public app directory yet, so you can't add it from a fresh Make account by searching for it today. The same transcription runs through the n8n community node or the REST API in the meantime.
Why does it use webhooks instead of just polling?
Make stops any module after 40 seconds, and transcription takes minutes — so a module that just waited for the result would time out. The instant webhook trigger fires the moment a transcript finishes, with no waiting inside a module at all; the polling trigger is the reliable fallback for scenarios that don't use webhooks.
What can I automate with it?
Watch for finished transcripts (instant or polling), transcribe a public video link or a file from an earlier module, then get, download or delete a transcript, list them, or call any endpoint directly with the universal module.
Which video platforms does it support?
Public YouTube, TikTok, Instagram, Facebook, Pinterest and Google Drive links, plus any file you already have in the scenario — including WhatsApp .opus voice notes.
Is it free?
The first transcript is free with no account (up to 40 minutes); after that, a free account raises the limit to 150 minutes and 500 MB per file — unlimited transcripts during the beta, no card needed.
Prefer a different tool? See the Zapier integration, the n8n node and REST API, or connect an AI assistant directly through MCP.