YouTube Video Summarizer

Created by AI Afterdark - Building Innovation with AI at Night An AI-powered application that generates summaries of YouTube videos and enables interactive conversations about their content.

Live Demo

Experience the YouTube Video Summarizer in action: https://aiafterdark-youtube-summarizer.streamlit.app/

Features

• Robust YouTube video transcript extraction with multiple fallback methods:
  • YouTube Transcript API (primary)
  • Pytube captions
  • yt-dlp caption extraction
• AI-powered content summarization using OpenRouter's LLMs
• Interactive Q&A about video content
• Adjustable summary detail levels
• Clean, responsive UI
• Comprehensive error handling and reporting

Cloud Deployment (Default)

Prerequisites

• Python 3.11+
• pip (Python package manager)
• Git
• OpenRouter API key

Quick Start

1. Clone the Repository

git clone https://github.com/AIAfterDark/youtube-summarizer-app.git
cd youtube-summarizer-app

2. Set Up Virtual Environment

python -m venv venv
# Windows
venv\Scripts\activate
# Unix/MacOS
source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

4. Configure Environment Create a .env file in the root directory and add your OpenRouter API key:

OPENROUTER_API_KEY=your_api_key_here

5. Run the Application

streamlit run app.py

Local Deployment (Ollama)

The app-local.py version allows you to run the summarizer using Ollama on your local machine, which is free and doesn't require an API key.

Prerequisites

• All requirements from Cloud Deployment
• Ollama installed on your machine

Setup Steps

1. Install Ollama

   • Download from ollama.ai
   • Follow the installation instructions for your OS
   • Make sure Ollama is running in the background

2. Pull Your Preferred Model

# Pull the default model (recommended)
ollama pull llama2

# Or pull other supported models
ollama pull codellama
ollama pull mistral
ollama pull neural-chat

3. Run the Local Version

streamlit run app-local.py

Available Local Models

The following models are tested and supported in app-local.py:

• llama2 (default, recommended)
• codellama
• mistral
• neural-chat

Configuration

You can modify these settings in app-local.py:

• Default model: Change model="llama2" in the ollama_completion function
• API endpoint: Default is http://localhost:11434/api/chat
• Timeout settings: Default is 30 seconds

Configuration

Summary Detail Level

Adjust the chunk size based on video length:

• Short videos (<30 mins): 4000
• Long content (1hr+): 7000+

Cloud Models (app.py)

The app uses OpenRouter's API to access various LLM models:

• meta-llama/llama-2-13b-chat (default)
• anthropic/claude-2
• openai/gpt-3.5-turbo

Contributing

We welcome contributions! Please feel free to submit a Pull Request.

License

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

Credits

Created by Will at AI Afterdark Built using:

• Streamlit for web interface
• OpenRouter for cloud AI
• Ollama for local AI
• YouTube Transcript API for content extraction

Contact

• Twitter: @AIAfterdark
• GitHub: AI Afterdark

---

Built by AI Afterdark - Innovating with AI at Night

YouTube Video Summarizer

An AI-powered Streamlit app that generates summaries of YouTube videos and allows you to chat with the content. Available in two versions: Cloud (OpenRouter) and Local (Ollama).

Features

• YouTube video transcript extraction
• Intelligent text chunking and processing
• AI-powered summarization
• Interactive chat with video content
• Multiple transcript retrieval methods
• Cloud and Local deployment options

Versions

Cloud Version (app.py)

• Uses OpenRouter API for AI inference
• Requires OpenRouter API key
• Better for deployment and sharing
• Uses meta-llama/llama-3.2-3b-instruct:free model

Local Version (app-local.py)

• Uses Ollama for local AI inference
• No API key required
• Better for privacy and offline use
• Supports multiple Ollama models

Setup

1. Clone the repository:

git clone https://github.com/yourusername/youtube-summarizer-app.git
cd youtube-summarizer-app

2. Install dependencies:

pip install -r requirements.txt

3. Setup based on version:

For Cloud Version (app.py):

1. Get an API key from OpenRouter
2. Create a .env file:

OPENROUTER_API_KEY=your_api_key_here

3. Run the app:

streamlit run app.py

For Local Version (app-local.py):

1. Install Ollama from ollama.ai
2. Pull the Llama2 model:

ollama pull llama2

3. Start Ollama:

ollama serve

4. Run the app:

streamlit run app-local.py

Usage

1. Enter a YouTube URL
2. Adjust the Summary Detail Level slider (1000-10000)
3. Click "Generate Summary"
4. View the generated summary
5. Use the chat to ask questions about the video content

Features in Detail

Transcript Retrieval

• Primary: YouTube Transcript API
• Fallback: yt-dlp
• Supports both manual and auto-generated captions

Text Processing

• Smart chunking based on sentence boundaries
• Context-aware summarization
• Clean transcript formatting

Chat Interface

• Context-aware responses
• Strictly based on generated summary
• Clear indication when information isn't available

Requirements

• Python 3.7+
• Streamlit
• youtube-transcript-api
• yt-dlp
• For Cloud Version: OpenRouter API key
• For Local Version: Ollama

Contributing

Feel free to open issues or submit pull requests with improvements.

License

MIT License - feel free to use this project as you wish.