🟢 Beginner ⚙️ Type: Voice Assistant / Local Desktop AI 💸 Free & Open Source ⭐ 200+ GitHub Stars
What is Jarvis AI?
Jarvis (created by GitHub user alpunlu12-commits) is a lightweight, open-source desktop voice assistant inspired by Iron Man’s famous AI. It allows you to talk to your computer naturally using your microphone and receive intelligent spoken responses back.
Instead of relying on heavy local language models that require expensive graphics cards, this tool routes your voice commands through Google’s extremely fast and completely free Gemini API. This means it acts as a very smart, completely free alternative to built-in operating system assistants like Siri or Cortana.
The application runs quietly in the background on your terminal. You can activate it at any time by simply saying the wake word “Jarvis,” or you can type directly into its text box if you don’t want to speak out loud.
Who is it for?
- Marvel fans and tinkerers who want the cool factor of having a personal voice assistant named Jarvis running on their computer.
- Developers and students looking for a beginner-friendly Python project to learn how voice recognition and the Gemini API work together.
- Mac and Windows users who want a highly capable, hands-free AI assistant without paying for ChatGPT Plus voice features.
- Turkish speakers (the original repository and video tutorials are presented in Turkish, making it highly accessible for that community, though it works universally).
What makes it special?
- Powered by Gemini 1.5 — It utilizes Google’s state-of-the-art AI model, which means it is incredibly smart, fast, and completely free to operate via Google AI Studio.
- Automated Setup Script — You don’t need to be a Python expert. It includes a
setup.shscript that automatically downloads necessary audio libraries (like PortAudio) and configures your virtual environment. - Voice and Text Hybrid — You aren’t forced to use the microphone. You can seamlessly switch between speaking the wake word “Jarvis” or typing commands into the UI window.
- Quick Mute Shortcuts — Built-in privacy controls allow you to instantly cut the microphone by pressing
F4(Windows) orCmd+M(Mac).
Requirements before you start
Before installing Jarvis, ensure your computer is ready with these basic software requirements:
- Python 3 — Required to run the application code. (Usually pre-installed on macOS).
- Visual Studio Code (VS Code) — The easiest code editor to run the background terminal.
- Homebrew (Mac users only) — A package manager needed to install audio drivers.
- A Free Google Gemini API Key — You will need to generate a key from Google AI Studio.
- A Working Microphone — Required for the voice recognition features to hear your commands.
Step-by-step installation
Step 1 — Download the Repository
Download the Jarvis folder from GitHub to your computer. Open VS Code, and drag and drop the jarvis folder directly into the VS Code window (or go to File → Open Folder).
Step 2 — Install Homebrew (Mac Only)
If you are on a Mac and do not have Homebrew installed, open the VS Code terminal (Ctrl + `) and paste this command to install it:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"Once finished, close that terminal tab and open a fresh one.
Step 3 — Run the Automated Setup
In your VS Code terminal, execute the setup script. This will automatically install PortAudio (for the microphone), create a Python sandbox, and download all required packages:
bash setup.shStep 4 — Get your Gemini API Key
- Go to aistudio.google.com.
- Click on Get API Key in the left menu, then click Create API Key.
- Copy the long string of text generated.
Step 5 — Start Jarvis
When the setup script from Step 3 finishes, it will ask “Şimdi başlatılsın mı? (e/h)” which means “Start now? (y/n)”. Type e and press enter.
A window will pop up asking for your API key. Paste the Gemini key you copied in Step 4. Jarvis is now listening! Say “Jarvis” to wake him up.
💡 Tip for Future Runs: To start Jarvis the next day, simply open the VS Code terminal in the folder and type:
source venv/bin/activatepython main.py
Common errors and fixes
| Error | What it means | How to fix it |
|---|---|---|
pyaudio / PortAudio missing | The Python audio library could not find your system’s microphone drivers. | If the setup.sh script failed, you must install PortAudio manually. On Mac, run: brew install portaudio. On Windows, you may need to install the PyAudio wheel manually. |
Invalid API Key | The Gemini API rejected your connection. | Ensure you copied the entire key from Google AI Studio without any extra spaces. Generate a new key and replace the old one in the application settings. |
| Jarvis is not responding to voice | The microphone is muted, or the wake word wasn’t detected clearly. | Ensure you haven’t accidentally pressed the mute shortcut (F4/Cmd+M). Speak the word “Jarvis” clearly into your primary microphone. |
Free vs Paid comparison
| Feature | Jarvis AI (Free Open Source) | ChatGPT Plus (Voice Mode) |
|---|---|---|
| Monthly Cost | $0 (Using Gemini Free API) | $20 per month |
| Always-Listening Wake Word | ✅ Yes (“Jarvis”) | ❌ No (Requires clicking the app) |
| Voice Realism & Emotion | 🟡 Basic text-to-speech output | 🟢 Extremely realistic, emotive voices |
| Customizability | ✅ Complete access to Python code | ❌ Closed ecosystem |
Bottom line: If you want a fun, customizable, and completely free desktop assistant that actively listens for a wake word while you work, this Jarvis clone is a great lightweight option. However, if you want deeply emotional, conversational voice interactions without touching any code, ChatGPT’s paid app is superior.
Alternatives — 3 similar tools
1. Leon AI
An incredibly robust, open-source personal assistant built to live on your server. It focuses heavily on privacy, allowing you to run voice interactions entirely offline without any Google APIs. It is much more complex to set up than Jarvis but offers deeper system integrations.
2. Open Interpreter
An advanced terminal-based assistant that can execute code and control your computer. While not strictly a voice assistant out-of-the-box, it represents the bleeding edge of local desktop AI, allowing the assistant to open apps, change settings, and automate tasks.
3. ChatGPT Desktop App
The official desktop application from OpenAI. While the advanced voice features require a premium subscription, it provides the most polished and seamless voice chat experience available on macOS and Windows today without any command-line setup.
🚀 Want more free AI tools like this?
We find, test, and write setup guides for the best free and open-source AI tools — so you don’t have to dig through GitHub yourself.Browse Free AI Tools at globalaiforce.com/shop →
📸 Follow us for daily AI tool tips and tutorials: instagram.com/globalaiforce