Advanced AI-Powered YouTube Automation Pipeline 🔥 ⚙️

A comprehensive multi-model pipeline that automates YouTube content creation using the latest AI technologies.

🌟 Features

1. Multi-Model Prompt Generation

• Uses Gemini 2.0 Flash and LLama 3 for creative, diverse, and multi-layered prompt generation
• Generates multiple prompts with divergent storytelling elements (e.g., different character arcs, alternate endings)
• Includes dynamic genre-based prompts (e.g., Sci-fi, Fantasy, Thriller)

2. Multi-Modal Story and Image Generation

• Uses Gemini 2.0 Flash for text-to-image generation
• Adds Stable Diffusion XL (SDXL) to generate highly detailed and realistic supporting images
• Incorporates ControlNet for precise image consistency, ensuring continuity across frames

3. Hyper-Realistic Image Animation

• Uses AnimateDiff and EBSynth for multi-frame animation, creating seamless character motion and facial expressions
• Implements frame interpolation with RIFE (Real-Time Intermediate Flow Estimation) for ultra-smooth transitions
• Adds depth mapping for parallax effects, making animations feel more dynamic and lifelike

4. AI Voiceover with Emotion and Dynamic Effects

• Uses XTTS and Bark AI for multi-speaker, emotion-rich voiceovers
• Applies Real-Time Audio Effects (reverb, pitch variation) for cinematic depth
• Auto-syncs voiceover with animation timing

5. Advanced Video Compilation with Dynamic Effects

• Uses FFmpeg + MoviePy for frame-perfect sequencing
• Adds AI-generated transitions and special effects (e.g., motion blur, dynamic lighting)
• Implements dynamic zoom and pan effects for enhanced visual storytelling

6. Google Drive Storage and Metadata Generation

• Automatically saves the video, images, and text in a unique Google Drive folder
• Generates SEO-optimized metadata:
  • AI-generated title, description, and tags
  • Includes relevant keywords and hashtags
  • Creates a thumbnail image using Gemini + SDXL for visually appealing clickbait

🚀 Getting Started in Google Colab

Option 1: Use the Quick Setup Notebook

1. Create a new Google Colab notebook
2. Add and run the following code to set up the pipeline:

# Clone the repository
!git clone https://github.com/yourusername/ytpipline.git
%cd ytpipline

# Install dependencies
!pip install -r requirements.txt

# Launch the web interface
!python webapp.py --share

Option 2: Manual Setup in Colab

1. Create a new cell and run:

# Install dependencies
!pip install google-generativeai>=0.3.0 transformers>=4.30.0 torch>=2.0.0 torchvision>=0.15.0 diffusers>=0.21.0 accelerate>=0.21.0 moviepy>=1.0.3 tensorflow>=2.12.0 opencv-python>=4.7.0 ftfy>=6.1.1 scipy>=1.10.1 tqdm>=4.65.0 pydrive>=1.3.1 gradio>=3.36.1 numpy>=1.24.3 Pillow>=9.5.0 matplotlib>=3.7.2 python-dotenv>=1.0.0 omegaconf>=2.3.0 einops>=0.6.1 pyngrok>=6.0.0 ffmpeg-python>=0.2.0 google-auth>=2.22.0 google-auth-oauthlib>=1.0.0 google-auth-httplib2>=0.1.0 google-api-python-client>=2.97.0 PyYAML>=6.0.1

2. Upload all the project files to Colab or copy them using magic commands:

%%writefile youtube_automation_pipeline.py
# Paste the contents of youtube_automation_pipeline.py here

3. Run the web interface:

!python webapp.py --share

💻 Local Installation

1. Clone the repository:

git clone https://github.com/yourusername/ytpipline.git
cd ytpipline

2. Install dependencies:

pip install -r requirements.txt

3. Configure API keys: Create a .env file with your API keys:

GEMINI_API_KEY=your_gemini_api_key
LLM_API_KEY=your_llm_api_key
GOOGLE_DRIVE_API_KEY=your_google_drive_api_key

4. Run the web interface:

python webapp.py

📖 Usage

Web Interface

The web interface provides an easy way to use the pipeline:

1. Enter a theme for your video (e.g., "space exploration", "underwater world")
2. Choose a visual style (e.g., "cyberpunk", "watercolor", "cinematic")
3. Set your desired video duration
4. Click "Generate YouTube Video"
5. Wait for the pipeline to complete

Command Line

You can also run the pipeline from the command line:

python youtube_automation_pipeline.py --theme "space exploration" --style "cinematic" --duration 60

⚙️ Configuration

The pipeline is configured using config.yaml. Key settings include:

• prompt_generator: Settings for the creative prompt generation
• image_generator: Settings for image generation models
• animator: Animation settings including FPS and duration
• voiceover: Voice generation settings
• video_compiler: Video compilation settings
• drive_uploader: Google Drive integration settings

🔄 Pipeline Flow

1. Prompt Generation: Creates multiple creative story concepts
2. Image Generation: Generates visual scenes based on prompts
3. Animation: Animates static images into dynamic sequences
4. Voiceover: Creates narration synchronized with animations
5. Video Compilation: Combines animations and audio into a final video
6. Upload & Metadata: Uploads to Google Drive with SEO metadata

📝 Notes for Google Colab Users

• The pipeline is designed to work optimally in Google Colab's environment with GPU acceleration
• When using in Colab, enable GPU runtime for faster processing
• The shared link feature allows you to access the web interface from any device
• Large models like Stable Diffusion XL may require high-memory Colab instances

🔑 API Keys

To use all features, you'll need:

• Gemini API Key: For prompt and image generation
• Google Drive API Credentials: For uploading content (optional)

📋 Dependencies

See requirements.txt for a full list of dependencies.

📄 License

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