N8N: Auto-Import YouTube Transcripts to Notion Database

By Mihail Lungu, Founder | February 5, 2026 | 9 min read

What if every YouTube video you saved automatically appeared in your Notion workspace with a full transcript, summary, and metadata—without lifting a finger? This N8N workflow makes it reality.

The Content Organization Problem Every Creator Faces

You bookmark a YouTube video. Then another. And another. Within weeks, you have hundreds of "saved for later" videos scattered across YouTube playlists, browser bookmarks, and half-forgotten tabs. Sound familiar?

The real problem isn't saving videos—it's accessing the knowledge inside them. That 45-minute tutorial on advanced Python techniques? Buried somewhere in your "Watch Later" graveyard. The brilliant marketing breakdown from that conference talk? Good luck finding the exact quote you need.

Content creators, researchers, and knowledge workers face this daily:

• No searchability — You can't search inside video content
• Time sink — Rewatching to find a specific section wastes hours
• Context lost — Why did you save this video again?
• Platform lock-in — Your notes live separate from the source

The Automated Solution: N8N + Scriptube + Notion

This workflow transforms how you consume and organize video content. Here's what happens automatically:

1. Trigger — Save a YouTube URL (via form, Slack command, or webhook)
2. Extract — Scriptube API pulls the complete transcript in seconds
3. Process — GPT-4 generates a summary, key points, and tags
4. Store — Everything lands in your Notion database, fully searchable

The result? A searchable video knowledge base that grows automatically. Find any quote, concept, or tutorial in seconds instead of rewatching entire videos.

Complete N8N Workflow Setup

Let's build this step by step. You'll need:

• N8N instance (cloud or self-hosted)
• Scriptube API key (free tier works for testing)
• Notion integration token
• OpenAI API key (optional, for AI summaries)

Step 1: Create the Webhook Trigger

Start with a Webhook node to receive YouTube URLs. This lets you trigger the workflow from anywhere—Slack, a Chrome extension, or a simple form.

{
  "nodes": [
    {
      "name": "Webhook",
      "type": "n8n-nodes-base.webhook",
      "parameters": {
        "path": "youtube-to-notion",
        "httpMethod": "POST"
      }
    }
  ]
}

Step 2: Fetch Transcript via Scriptube API

Add an HTTP Request node to call Scriptube's transcript endpoint:

{
  "name": "Get Transcript",
  "type": "n8n-nodes-base.httpRequest",
  "parameters": {
    "method": "POST",
    "url": "https://api.scriptube.io/v1/transcript",
    "headers": {
      "Authorization": "Bearer {{ $credentials.scriptubeApiKey }}",
      "Content-Type": "application/json"
    },
    "body": {
      "url": "{{ $json.youtube_url }}",
      "format": "text",
      "include_timestamps": true
    }
  }
}

Scriptube handles all the heavy lifting: extracting captions, handling different caption formats, and returning clean text ready for processing.

Step 3: Generate AI Summary (Optional but Powerful)

Add an OpenAI node to create intelligent summaries:

{
  "name": "Summarize with GPT-4",
  "type": "n8n-nodes-base.openAi",
  "parameters": {
    "model": "gpt-4-turbo",
    "messages": [
      {
        "role": "system",
        "content": "You are a content analyst. Summarize video transcripts with: 1) One-paragraph summary, 2) 5 key takeaways as bullet points, 3) Relevant tags (comma-separated)"
      },
      {
        "role": "user",
        "content": "Summarize this transcript:\n\n{{ $json.transcript }}"
      }
    ]
  }
}

Step 4: Create Notion Page

Finally, push everything to Notion with full metadata:

{
  "name": "Create Notion Page",
  "type": "n8n-nodes-base.notion",
  "parameters": {
    "databaseId": "{{ $credentials.notionDatabaseId }}",
    "properties": {
      "Title": { "title": [{ "text": { "content": "{{ $json.video_title }}" }}] },
      "URL": { "url": "{{ $json.youtube_url }}" },
      "Channel": { "rich_text": [{ "text": { "content": "{{ $json.channel_name }}" }}] },
      "Duration": { "number": {{ $json.duration_minutes }} },
      "Tags": { "multi_select": {{ $json.tags }} },
      "Added": { "date": { "start": "{{ $now.format('YYYY-MM-DD') }}" }}
    },
    "children": [
      {
        "type": "heading_2",
        "heading_2": { "rich_text": [{ "text": { "content": "Summary" }}] }
      },
      {
        "type": "paragraph",
        "paragraph": { "rich_text": [{ "text": { "content": "{{ $json.summary }}" }}] }
      },
      {
        "type": "heading_2",
        "heading_2": { "rich_text": [{ "text": { "content": "Key Takeaways" }}] }
      },
      {
        "type": "bulleted_list_item",
        "bulleted_list_item": { "rich_text": [{ "text": { "content": "{{ $json.key_points }}" }}] }
      },
      {
        "type": "heading_2",
        "heading_2": { "rich_text": [{ "text": { "content": "Full Transcript" }}] }
      },
      {
        "type": "paragraph",
        "paragraph": { "rich_text": [{ "text": { "content": "{{ $json.transcript }}" }}] }
      }
    ]
  }
}

Notion Database Configuration

Create a Notion database with these properties for maximum usefulness:

Pro tip: Add a "Transcript Search" property as a formula that concatenates key fields. Notion's search will index it, making full-text transcript search possible.

Bonus: Multi-Language Support

Scriptube supports transcript extraction in 50+ languages and can translate on the fly. Add a language parameter to your workflow:

{
  "body": {
    "url": "{{ $json.youtube_url }}",
    "translate_to": "en",
    "source_language": "auto"
  }
}

Now you can build a knowledge base from any YouTube content, regardless of original language. Japanese tech tutorials, Spanish marketing webinars, German engineering lectures—all automatically transcribed and translated.

Convert Transcripts to Audio with ElevenLabs

Want to listen to transcripts instead of reading? Add an ElevenLabs node to generate natural-sounding audio:

{
  "name": "Generate Audio",
  "type": "n8n-nodes-base.httpRequest",
  "parameters": {
    "method": "POST",
    "url": "https://api.elevenlabs.io/v1/text-to-speech/{{ $credentials.voiceId }}",
    "headers": {
      "xi-api-key": "{{ $credentials.elevenLabsKey }}"
    },
    "body": {
      "text": "{{ $json.summary }}",
      "model_id": "eleven_turbo_v2"
    }
  }
}

This creates a podcast-style audio version you can listen to during commutes—turning video content into audio content automatically.

Real Results & ROI

Users running this workflow report significant time savings:

• 80% reduction in time spent searching for video content
• 10x faster content consumption (reading vs watching)
• Zero manual data entry — everything populates automatically
• 100% recall — every video becomes permanently searchable

A content creator processing 50 videos per week saves approximately 15 hours monthly on research and organization alone.

Cost Breakdown

For the time saved, that's roughly $3/hour invested in automation—far below minimum wage for manual work.

Troubleshooting Common Issues

Transcript Not Found

Some videos have disabled captions. Scriptube can use Whisper AI to generate transcripts from audio—enable the force_whisper parameter for these cases.

Notion Rate Limits

Notion's API allows 3 requests/second. Add a "Wait" node between batches if processing many videos at once.

Long Transcripts Truncated

Notion blocks have character limits. Split long transcripts into multiple blocks using N8N's "Split In Batches" node.

Keep Reading

• N8N: AI-Summarize YouTube Transcripts with GPT-4
• N8N: Build Research Database from YouTube Channels
• Speed Learning: Skim Transcripts Instead of Watching Videos