- Overview
- Key Features
- Technologies Used
- Installation
- Usage
- Demo
- Contributing
- License
- Acknowledgments
VidSense is a browser extension that provides instant answers and concise summaries based on the YouTube video you're watching. It offers AI-powered responses directly within your current tab, enhancing your video-watching experience.
VidSense uses YouTube video captions to provide accurate and contextually relevant answers to user queries. This feature allows users to quickly find precise information without watching the entire video.
Get concise summaries of YouTube videos to grasp the main points and essential information quickly. This feature helps users efficiently access key ideas without watching the full content.
The question-answering tab also supports any other prompts such as "Generate a quiz" or "Translate to Hindi", further enhancing the utility of the extension.
- Frontend: HTML/CSS/JavaScript
- Backend: Flask
- AI Integration: Google Gemini
- Containerization: Docker
- API Integration: MindsDB
- Download the latest release here.
- Extract the files to your preferred location.
-
Install Docker Desktop.
-
Set up MindsDB on Docker following this guide.
-
Open the SQL editor in the MindsDB extension and run the following queries:
CREATE ML_ENGINE google_gemini_engine FROM google_gemini USING google_gemini_api_key = 'your-api-key'; CREATE MODEL google_gemini_model PREDICT answer USING engine = 'google_gemini_engine', column = 'question', model = 'gemini-pro';
-
Download and install Python. Skip this step if it is already installed on your computer.
-
Run the
install-dependencies.bat
file in the extracted folder.
- Enable Extension developer mode in your browser.
- Click on "Load Unpacked" and select the extension folder from the extracted files.
- The VidSense extension will now be available in your browser's extensions menu.
- Make sure the MindsDB Docker container is running
- Start the server by running
start.bat
. - Navigate to a YouTube video.
- Click the VidSense extension icon to open the interface.
- Use the summary or question-answering features as needed.
- Stop the server using
stop.bat
when finished.
(Click to watch on YouTube)
If you'd like to contribute, please follow these steps:
-
Fork the repository
-
Create your feature branch:
git checkout -b your-name/feature
-
Commit your changes:
git commit -m 'Add some feature'
-
Push to the branch:
git push origin your-name/feature
-
Open a Pull Request
This project is licensed under the MIT License.
- MindsDB for AI Data automation
- Google Gemini API for LLM tasks
- Flask for the backend framework
- Docker for containerization
- youtube-transcript-api for extracting YouTube captions.