Automate YouTube Video Summaries & Transcripts with n8n

What This Workflow Does

This workflow takes a YouTube video URL and sends it to Google’s Gemini AI to get text outputs that help understand the video quickly.
It can make transcripts, summaries, scene details, or social media clips based on your choice.
This saves many hours of watching and writing by hand.

You input a video link and select what you want: full text, timestamps, summary, or scenes.
The workflow handles requests and returns useful text you can use fast.

Who Should Use This Workflow

This is good for anyone who works with video content and needs text results quickly.
Examples include bloggers, social media managers, teachers, and marketers.

If spending many hours typing transcripts or summaries wastes time, this workflow helps a lot.
It also fits users who want fast text info for video analysis without heavy manual work.

Tools and Services Used

• n8n Automation Platform: to build and run the workflow.
• Google Gemini AI API: for generating text content from videos.
• YouTube: source of public videos for analysis.

The workflow connects these to automate video text extraction smoothly.

Inputs, Processing, and Outputs

Inputs

• YouTube video URL (must be public).
• API Key for Google Gemini AI.
• Prompt type indicating desired output (e.g. transcript, summary).

Processing

• Workflow picks prompt type and sets the right AI prompt.
• Sends a detailed request to Google Gemini’s API including video URL.
• Receives AI-generated text results and merges them with existing data.
• Prepares clean outputs including text and token usage statistics.

Outputs

• Text transcript or summary or scene description based on selection.
• Token counts and AI model version for monitoring.

Beginner Step-by-Step: Using This Workflow in Production

Import Workflow

1. Download the workflow file using the Download button on this page.
2. Open the n8n editor where all workflows are managed.
3. Click on “Import from File” and choose the downloaded workflow file.

Configure Settings

1. Find the Set: Define Initial Variables node.
2. Enter your Google API Key exactly, and paste the YouTube video URL to analyze.
3. Choose prompt type from: transcript, timestamps, summary, scene, clips.
4. Update any IDs, emails, or folder names if the workflow uses external services.

Test and Activate

1. Run the workflow once using the Manual Trigger node to check outputs.
2. Look at the output, fix any errors or wrong inputs.
3. When satisfied, activate the workflow by toggling the active switch in n8n.

If self-hosting n8n, see self-host n8n for deployment tips.

Workflow Breakdown

Step 1: Manual Trigger

This node starts the process when clicked manually.
Useful for testing or on-demand runs.

Step 2: Set Initial Variables

This sets key info like API Key, video URL, and prompt type.
These feed into later nodes.

Step 3: Switch Node

Switch chooses which branch to run based on prompt type.
Routes the workflow to desired output format.

Step 4: Set Prompt Nodes

Each output type has its own Set node with a special text prompt.
The prompt guides Gemini AI on what text to generate.

Step 5: Merge Data

A Set node merges previous info to prepare for API call.
Keeps all important data together.

Step 6: HTTP Request to Google Gemini

Sends a POST request with prompt and video URL.
Receives AI text results.

Step 7: Code Node Merges Output

This JavaScript merges AI response with prior node data.
Ensures no info is lost in transformation.

Step 8: Set Final Fields

This cleans the AI response into named fields.
Makes output easy to read and use.

Step 9: Optional Error Handling

If enabled, checks for API errors and handles them properly.
Helps keep workflow stable.

Customizations

• Change model names in Set nodes to test different AI versions.
• Use a Webhook node instead of Manual Trigger for automatic runs.
• Connect output to Notion, Google Docs, or Airtable to save text results.
• Send multiple prompts in one run by duplicating branches from the Switch node.
• Edit prompt text to fit specific video types or add SEO keywords.

Troubleshooting

• Unauthorized Errors: Check API Key is correct and active in Google Cloud Console.
• Empty or Wrong Output: Confirm YouTube URL is public and correct.
• Code Node Reference Errors: Confirm exact node names in JavaScript code.

Pre-Production Checklist

• Verify API Key is valid and not expired.
• Ensure YouTube video URL is publicly accessible.
• Check all prompt texts are correct and complete.
• Run tests for all prompt types manually.
• Backup workflow before changes.

Deployment Guide

Turn the workflow active in n8n.
If using scheduled runs or webhooks, set triggers accordingly.

Watch execution logs to confirm correct responses.
Add notifications or saving actions after processing if needed.

Conclusion

This workflow cuts down manual video text work by using Google Gemini AI inside n8n.
Inputting a YouTube link and selecting output type gives fast transcripts or summaries.

It saves time, increases productivity, and reduces errors.
Users get clear, ready text they can use immediately.

Try importing and testing the workflow to see how much manual effort can be removed.

Summary

✓ Saves hours of manual video transcription and summarization.
✓ Lets user choose type of output text from one video link.
✓ Sends clear prompts to Google Gemini AI for precise results.
✓ Runs inside n8n for easy automation and extension.
✓ Works well for content creators, marketers, and educators who handle video text.