N8N Video Processing Code Nodes

A collection of custom code nodes for n8n to enhance video processing workflows. These nodes provide advanced video analysis, frame extraction, and metadata generation capabilities.

Components

1. Best Frame Extractor

Extracts 5 key frames from a video and selects the best one for social media using OpenAI's GPT-4o Vision.

Features:

• Scene detection for intelligent frame selection
• Visual appeal and composition analysis
• Social media platform-specific suggestions
• Detailed metadata for each extracted frame

Files:

• best-frame-extractor/best-frame-extractor.js - Main code node script
• best-frame-extractor/configuration.md - Configuration documentation

2. Video Context Extractor

Extracts comprehensive context from videos including transcripts, visual analysis, and topic segmentation.

Features:

• Word-level timestamp transcription using Whisper API
• Visual frame analysis with GPT-4o Vision
• Automatic topic segmentation with timestamps
• Combined transcript and visual context

Files:

• video-context-extractor/video-context-extractor.js - Main code node script
• video-context-extractor/configuration.md - Configuration documentation

3. Docker Configuration

Docker setup for n8n with ffmpeg and other required dependencies pre-installed.

Files:

• docker-config/docker-compose.yml - Docker Compose configuration
• docker-config/n8n.Dockerfile - Custom Dockerfile for n8n with ffmpeg

Requirements

• n8n v1.91.1 or later
• Docker and Docker Compose
• FFmpeg installed in the container
• OpenAI API key

Setup Instructions

1. Clone this repository
2. Navigate to the repository directory
3. Start the Docker container:

   cd docker-config
   docker-compose up -d
   

4. Access n8n at http://localhost:5678
5. Create a new workflow and add a "Code" node
6. Copy the content of the desired script into the Code node
7. Configure the workflow as described in the configuration documentation

Environment Variables

The following environment variables are recommended for optimal performance:

N8N_DEFAULT_BINARY_DATA_MODE=filesystem
N8N_BINARY_DATA_STORAGE_PATH=/tmp/n8n-binary-data
N8N_PAYLOAD_SIZE_MAX=1024

License

MIT

Author

Created for video processing workflows in n8n.