Liminal.ai

A content-to-video conversion pipeline that transforms longform text or PDFs into short-form videos.

Overview

This project transforms longform text content into engaging short-form videos through a series of AI-powered steps:

1. Content Chunking: Breaks down articles into script segments optimized for short videos
2. Text-to-Speech: Converts scripts into natural-sounding voiceovers
3. B-Roll Generation: Creates relevant visual content to accompany the narration
4. Video Assembly: Combines all elements into the final video

Pipeline Components

• Chunker: Breaks down longform content into manageable chunks
• Script Generator: Creates a script from content chunks
• TTS Engine: Generates voiceover audio from the script
• B-Roll Generator: Creates visual content based on the script
• Video Assembler: Combines audio and visuals into the final video
• Orchestrator: Coordinates the entire pipeline process

Features

• Process articles, blog posts, or any longform text
• Generate engaging scripts with hooks and calls-to-action
• Create natural-sounding voiceovers using ElevenLabs
• Automatically assemble complete videos ready for social media

Installation

# Clone the repository
git clone https://github.com/yourusername/liminal-ai.git
cd liminal-ai

# Create a virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Configuration

Create a .env file in the root directory with your API keys:

ELEVENLABS_API_KEY=your_elevenlabs_api_key
ELEVENLABS_VOICE_ID=your_preferred_voice_id
ELEVENLABS_MODEL_ID=eleven_multilingual_v2
DEEPSEEK_API_KEY=your_deepseek_api_key

Usage

from pipeline.orchestrator import process

# Process an article
article_text = """
Your article text goes here...
"""

process(article_text)

Project Structure

liminal-ai/
├── pipeline/
│   ├── __init__.py
│   ├── chunker.py
│   ├── tts.py
│   ├── broll.py (coming soon)
│   ├── assembler.py (coming soon)
│   └── orchestrator.py
├── output/
│   ├── audio/
│   ├── video/
│   └── scripts/
├── .env
├── requirements.txt
└── README.md

Requirements

• Python 3.8+
• ElevenLabs API key
• DeepSeek API key
• FFmpeg (for video processing)
• pydub
• requests
• python-dotenv
• beautifulsoup4
• nltk

Setup and Usage

[Instructions here]

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.