VidMind is the ultimate open-source tool for AI-powered content analysis. Built for students, researchers, and content creators, it offers a seamless experience to turn any media into a comprehensive learning experience. Its intuitive Graphical User Interface (GUI) makes AI-powered analysis accessible to everyone, regardless of technical expertise.
- Universal Input: Process content from YouTube, 1000+ websites, local audio/video files, and documents (
.pdf,.docx,.txt). - AI-Powered Analysis: Leverage the power of large language models to:
- Generate Summaries: Get both brief and detailed summaries of your content.
- Extract Key Moments: Identify the most important topics and timestamps.
- Create Learning Tools: Automatically generate flashcards, quizzes, and learning roadmaps, Q&A, Quizes, etc....
- Generate Q&A and FAQs: Ask questions and get answers from your content.
- Flexible AI Provider Support:
- Gemini: Access GEMINI 2.5 PRO/FLASH For Free
- OpenRouter: Access a wide variety of open-source and proprietary models. "Recommended - openai/gpt-3.5-turbo"
- Professional Export Options: Save your enhanced content as
.txt,.md,.pdf, and Anki-compatible.csvfiles For Easy ANKI import "Check Below For Steps To import into FREE OpenSource ANKI FLASHCARDs ". - Smart & Efficient:
- Remembers your preferred settings.
- Caches results to avoid reprocessing and save on API costs.
- 100% Free & Open-Source: This project is completely free and open-source under the MIT License.
# Clone the repository
git clone https://github.com/yomazini/VidMind.git
cd VidMind
# Run the setup script
chmod +x install.sh
./install.sh
# Start the application
streamlit run mediaGenius/app.py- Install Python: Ensure you have Python 3.12+ installed. This project is not stable on Python 3.10.
- Install Dependencies:
pip install -r mediaGenius/requirements.txt
- Install FFmpeg:
- Debian/Ubuntu:
sudo apt install ffmpeg - macOS (Homebrew):
brew install ffmpeg - Windows: Download from ffmpeg.org and add to your system's PATH.
- Debian/Ubuntu:
- Run the App:
streamlit run mediaGenius/app.py
➡️ Explore ScriptFlow And VidMind Youtube Tutorial
Need to transcribe a large volume of audio or video/playlist, from any website? Check out our Brother project, ScriptFlow.
ScriptFlow transforms any video or audio into searchable, editable, and exportable text with AI precision. It's the ultimate open-source tool for AI-powered transcription.
➡️ Explore ScriptFlow on GitHub
VidMind supports multiple AI providers. Here's what you need to know:
- Pros: Fast, reliable, and offers a generous free tier.
- How to get a key: Visit Google AI Studio.
- Pros: Access to a wide variety of models, including many free and open-source options.
- How to get a key: Visit OpenRouter.ai.
- Important Notes on the Free Tier:
- OpenRouter's free tier is great for testing, but has limitations.
- Daily Limits: Free models have strict daily limits (e.g., 50 requests/day).
- Model Availability: Community-hosted models may shut down after 24 hours of inactivity. You can use a "keep-alive" script to prevent this.
- Recommended Model: For a good balance of speed and reliability on the free tier, we recommend
openai/gpt-3.5-turbo.
To import your flashcards, open the Anki desktop application Or Phone Application "IOS or Android " and select the deck you want to add the cards to. Go to File > Import... and choose the .csv file you downloaded from VidMind. In the import window, ensure the Note Type is set to "Basic" and the correct deck is selected. The most important step is mapping the columns: make sure the first field from the CSV (your question) is mapped to the Front field in Anki, and the second field (your answer) is mapped to the Back field. Check that the field separator is set to "Tab" Or "Comma" based on what exported, then click "Import" to add your new flashcards to your deck. Just Make sure To add Question In front and Answer In Back;
| Model | Provider | Speed | Quality | Free Tier | Notes |
|---|---|---|---|---|---|
gemini-pro |
Gemini | Fast | High | Generous | May experience Delay On Gemini 2.5 Pro reasoning Model. |
openai/gpt-3.5-turbo |
OpenRouter | Fast | Good | Limited | Recommended Best balance on OpenRouter's free tier. |
mistral-7b-instruct |
OpenRouter | Medium | Good | Limited | Strong instruction-following model. |
This project is built with ❤️ by Youssef Mazini. If you find it useful, please consider showing your support.
- Star the repo on GitHub: 🌟 Star VidMind
- Connect on Social Media:
- LinkedIn: Youssef Mazini
- X (Twitter): @mazini_youssef
- Report an Issue: 🐛 Report a Bug
