Send voice notes to n8n with QuickCast

Last updated: 30 October 2026

QuickCast records a voice note and posts it to a webhook URL you choose. n8n’s Webhook node can receive that post, so one tap on your phone can start any n8n workflow: save the audio, file the transcript, or hand both to an AI step.

Checked on 30 October 2026 against n8n 2.45.7 (the official Docker image). We sent a request built to the format described in our privacy article to a local n8n and inspected what the Webhook node received. We did not send it from the phone app, and the steps below for a phone to reach your own n8n depend on your network.

Get QuickCast: App Store · Google Play

1. Create the webhook in n8n

  1. Add a Webhook node as the first node of a new workflow.
  2. Set HTTP Method to POST.
  3. Choose a Path, for example quickcast.
  4. Leave the binary options as they are. n8n names each uploaded file after its form field.

The node shows two URLs. The Test URL (/webhook-test/...) only works while the editor is waiting for a test event. The Production URL (/webhook/...) only works once the workflow is active. Use the Test URL while you build, then switch QuickCast to the Production URL.

2. Add the URL in QuickCast

Open QuickCast, add a webhook and paste the n8n URL. Keep the method on POST. Leave the audio field name on its default, audio, and the transcript field name on its default, transcript.

Your phone has to reach that address, and iOS requires a valid TLS certificate. So http://localhost:5678 will not work from a phone, and neither will a self-signed certificate. Use your n8n Cloud URL, a public domain with a normal certificate, or a private network such as Tailscale with its built-in HTTPS certificates (see the privacy article).

If your n8n instance needs a key, add it as a custom header in QuickCast and require the same header in the Webhook node’s authentication setting.

3. Record and send

Tap record, say your note and stop. The recording uploads when it finishes. In n8n, open Executions to see the run.

What n8n receives

This is what we saw on 30 October 2026 for an audio note with a transcript:

In the workflowWhat it holds
Binary property audioThe recording. fileName and mimeType come from the upload, for example audio/m4a
{{ $json.body.transcript }}The transcript text, when transcription is on and not empty

The audio is not in the JSON. Node settings that ask for a “binary property” take the name audio.

A text-only capture is sent as JSON, {"text": "..."}, with text as the default field name. In n8n that text arrives in {{ $json.body.text }}. We did not test this one in n8n.

Ideas for the next node

  • Save the transcript: pass {{ $json.body.transcript }} to a Google Sheets, Notion or Slack node. QuickCast transcribes on the phone, so you need no separate transcription step for that.
  • Keep the audio: send the audio binary property to a storage node such as Google Drive or S3.
  • Transcribe it again yourself: pass the audio binary property to the transcription node you prefer.
  • Send one note to several places: QuickCast can post the same recording to several webhooks at once, for example n8n plus a backup address.

When it does not work

  • 404 “webhook is not registered”: you used the Test URL without clicking Listen for test event, or you used the Production URL while the workflow is inactive.
  • The send fails on the phone: check the certificate, and check that the phone can open the URL in its browser. Open the item in QuickCast’s history to see the error. Failed sends are kept so you can resend them.
  • Big recordings are refused: n8n limits the request body size, 16 MB by default. A long recording in WAV or FLAC can pass that; AAC (M4A) is smaller.
  • Anyone could call the URL: since version 1.7 QuickCast can sign each request, so your server can check it came from you. See QuickCast 1.7.

QuickCast has no server of its own. The recording goes from your phone straight to your n8n.