This is a Python application that creates comprehensive summaries of YouTube videos using Groq's AI models. The application can either use YouTube's transcription or create its own using Groq's Whisper Large V3 Turbo when no transcript is available.
- Automatic transcript extraction from YouTube videos
- Fallback to audio transcription using Groq's Whisper Large V3 Turbo
- Advanced text chunking with Langchain
- Comprehensive summarization using Llama 3.1 8B Instant
- Multi-language support with 12+ languages
- Language selection for summaries
- Structured summaries with clear sections
- Clean and intuitive Streamlit web interface
- Progress tracking and status updates
Before you begin, ensure you have installed:
- Python 3.6 or above
- FFmpeg (required for audio processing)
- Clone this repository:
git clone https://github.com/DevRico003/youtube_summarizer- Change into the project directory:
cd youtube_summarizer- Install required packages:
pip install -r requirements.txt- Install FFmpeg (Ubuntu/Debian):
sudo apt-get update
sudo apt-get install ffmpegFor other operating systems, please refer to the FFmpeg installation guide
- Create a
.envfile in your project directory and add your Groq API Key:
GROQ_API_KEY=your_groq_api_key- Update the
env_pathvariable inapp.pyto match your.envfile location.
- Start the application:
streamlit run app.py-
Open your web browser to the provided URL (typically http://localhost:8501)
-
Enter a YouTube video URL in the input field
-
Select your desired summary language from the dropdown menu
-
Click "Generate Summary"
The application will:
- Attempt to fetch the YouTube transcript
- If no transcript is available, download and transcribe the audio using Groq's Whisper
- Process and chunk the text appropriately
- Generate a comprehensive summary with the following sections:
- 🎯 Title
- 📝 Overview
- 🔑 Key Points
- 💡 Main Takeaways
- 🔄 Context & Implications
Each summary includes:
- A descriptive title
- A brief overview (2-3 sentences)
- Key points with examples and data
- Practical insights and actionable conclusions
- Context and broader implications
The application includes robust error handling for:
- Invalid YouTube URLs
- Missing transcripts
- Failed audio downloads
- API errors
- Network issues
- Uses Groq's API with OpenAI compatibility layer
- Implements efficient text chunking with Langchain
- Uses Llama 3.1 8B Instant model for summarization
- Employs yt-dlp for reliable video processing
- Includes automatic cleanup of temporary files
- Features progress tracking and user feedback
The application supports summaries in multiple languages:
- English
- German (Deutsch)
- Spanish (Español)
- French (Français)
- Italian (Italiano)
- Dutch (Nederlands)
- Polish (Polski)
- Portuguese (Português)
- Japanese (日本語)
- Chinese (中文)
- Korean (한국어)
- Russian (Русский)
Simply select your preferred language from the dropdown menu, and the summary will be generated in that language, regardless of the original video language.
Contributions are welcome! Please feel free to submit a Pull Request.
Distributed under the MIT License. See LICENSE for more information.
This application uses Groq's API services. Usage will incur costs based on your Groq account. Please review Groq's pricing structure before extensive use.
To access YouTube transcripts, you need to provide authentication cookies. Follow these steps:
- Open the Chrome Web Store
- Search for "Get cookies.txt"
- Install the "Get cookies.txt LOCALLY" extension
- Click the extension icon to ensure it's pinned to your browser
- Go to YouTube
- Sign in to your YouTube/Google account
- Click the "Get cookies.txt" extension icon
- Click "Export" to download the cookies file
- Rename the downloaded file to
cookies.txt - Place the
cookies.txtfile in the same directory asapp.py - Ensure the file permissions are correct (readable by the application)
Note: Keep your cookies.txt file secure and never share it publicly, as it contains your authentication information.