Lang-Graph is an automation pipeline designed to scrape AI/Tech news, enhance articles using OpenAI, generate YouTube Shorts scripts, create audio narrations, and produce videos using Json2Video and FFmpeg.

• News Scraping: Scrapes AI/Tech news from multiple sources like TechCrunch, The Verge, and VentureBeat.
• Article Enhancement: Uses OpenAI to summarize, categorize, and score articles.
• Excel Report: Creates a structured Excel report of the articles.
• YouTube Shorts Script: Generates engaging scripts optimized for YouTube Shorts.
• Audio Narration: Creates audio using ElevenLabs or system TTS.
• Video Creation: Produces videos with synchronized subtitles and visuals using Json2Video and FFmpeg.

Set the following environment variables before running the project:

• OPENAI_API_KEY: Your OpenAI API key for article enhancement and script generation.
• ELEVENLABS_API_KEY: (Optional) ElevenLabs API key for audio generation.
• VIDEO_OUTPUT_DIR: Directory where output files (Excel, audio, video) will be saved.
• JSON2VIDEO_API_KEY: Json2Video API key for video creation.
• UNSPLASH_API_KEY: (Optional) Unsplash API key for image search.

Example:

export OPENAI_API_KEY="your-openai-api-key"
export ELEVENLABS_API_KEY="your-elevenlabs-api-key"
export VIDEO_OUTPUT_DIR="/path/to/output"
export JSON2VIDEO_API_KEY="your-json2video-api-key"
export UNSPLASH_API_KEY="your-unsplash-api-key"

1. Clone the Repository:

   git clone https://github.com/your-repo/lang-graph.git
   cd lang-graph

2. Set Up Virtual Environment:

   python3 -m venv .venv
   source .venv/bin/activate

3. Install Dependencies:

   pip install -r requirements.txt

4. Install FFmpeg:

   • Mac: brew install ffmpeg
   • Linux: sudo apt install ffmpeg
   • Windows: Download from FFmpeg.org.

If you prefer to run the pipeline directly without Uvicorn:

1. Run the Pipeline:

   python scripts/runner.py

2. Outputs:

   • Excel Report: Saved in the VIDEO_OUTPUT_DIR.
   • Audio File: Generated narration saved in the VIDEO_OUTPUT_DIR.
   • Video File: Final video saved in the VIDEO_OUTPUT_DIR.

The project can be run as an API using Uvicorn.

1. Install Uvicorn:

   pip install uvicorn

2. Run the API:

   uv run runner.py

• OpenAI: Get your API key.
• ElevenLabs: Get your API key.
• Json2Video: Get your API key.
• Unsplash: Get your API key.

export OPENAI_API_KEY="your-openai-api-key"
export ELEVENLABS_API_KEY="your-elevenlabs-api-key"
export VIDEO_OUTPUT_DIR="/path/to/output"
export JSON2VIDEO_API_KEY="your-json2video-api-key"
export UNSPLASH_API_KEY="your-unsplash-api-key"

• ElevenLabs: If not available, the pipeline will use system TTS or create silent audio files.
• Json2Video: Ensure your API key is valid for video creation.
• FFmpeg: Required for video merging and fallback video creation.

1. Missing API Keys: Ensure all required environment variables are set.

2. FFmpeg Not Found: Install FFmpeg and ensure it is added to your system's PATH.

3. Json2Video Errors: Verify the payload structure and API key validity.