← Documentation: Instagram Reel to Notion Second Brain
Apify retrieves the spoken transcript and available caption and author metadata. These are the source texts used by the AI stage; private, inaccessible or silent Reels may not provide usable results.
Use an Apify account with access to Instagram Video Transcript Scraper. Create an n8n Header Auth credential with name Authorization and value Bearer YOUR_APIFY_API_TOKEN.
Assign it to 08 — Start Transcript Run, 10 — Wait & Re-check Transcript Run and 11 — Read Transcript Result. A temporary authenticated GET to https://api.apify.com/v2/users/me can confirm the account connection.
The launch request uses:
POST https://api.apify.com/v2/acts/lance_api~instagram-video-transcript-scraper/runs
{
"urls": ["https://www.instagram.com/reel/SHORTCODE/"],
"formats": ["text"],
"translate": false,
"includeMediaMetadata": true
}
The workflow supplies the normalized URL dynamically. Keep includeMediaMetadata enabled to request caption and author details; their availability varies by Reel.
The start node waits with waitForFinish=60. If the run is still active, the workflow checks it again before retrieving dataset items. 12 — Validate Transcript extracts the usable text and handles empty or unsupported results.
Inspect the Apify run before retrying a timeout: the Actor may have continued processing. The launch includes maxTotalChargeUsd=1; review the Actor's support for this parameter and its pricing before regular use. Repeated submissions are processed again because no cache is enabled.
For authentication errors, check the Bearer prefix and credential assignment. For an unknown Actor, compare its slug with the configured endpoint.
For empty text, inspect dataset status, success and transcript.text. Missing content.caption is different from missing audio: the workflow can continue with a usable transcript alone.
Continue with 03 · OpenAI Model & Knowledge Extraction.