Automate Audio & Video Transcription with n8n (Whisper + GPT-4o to Notion)
Recordings pile up — calls, interviews, voice notes, webinars — and almost none of them get revisited because nobody has time to listen again. This n8n workflow turns any audio or video file into a clean transcript and an AI summary, then files it where you'll actually find it.
From recording to searchable knowledge
An hour of audio is unsearchable. A transcript with a summary is a knowledge asset you can grep, link, and act on. The workflow's job is to make that conversion automatic so every recording becomes useful text without manual effort.
Capturing the file
The trigger can be anything that drops a file: an upload form, a new file in Google Drive or Dropbox, or a message in Telegram. The workflow grabs the audio or video and, for video, extracts the audio track before transcription.
Skip the build — get the ready-to-import template
This is the transcription + summary workflow from the guide, prebuilt. Import it into n8n in minutes, connect your credentials, and go live today.
Get the n8n template on Gumroad →
Transcribing with Whisper
The audio goes to a Whisper-based transcription API, which returns accurate text with speaker timing. Whisper handles accents, background noise, and multiple languages well, which makes it a reliable default for real-world recordings rather than studio-clean audio.
Summarizing with GPT-4o
The raw transcript then goes to GPT-4o with a structured prompt: produce a short summary, a bulleted list of key points, and any action items or decisions. For meetings, asking explicitly for owners and due dates turns the summary into something the team can execute against.
Filing it where it belongs
The transcript and summary are written to a Notion database, emailed, or posted to Slack. Tag each entry by source or project and your recordings become a searchable archive instead of a folder of files nobody opens.
Ready to automate this?
You can wire every node above by hand — or import a tested workflow and be running in minutes. The template ships with the full n8n JSON, setup notes, and the exact prompts used in this guide.
