J.A.R.V.I.S. is a sleek, Python-based desktop voice assistant designed for extreme speed and utility. It uses a modern Push-to-Talk interface to instantly listen to your voice commands, process them locally or through an ultra-fast cloud AI, and speak the results back to you.
- Lightning Fast AI (Groq): Uses the
GroqAPI and thellama-3.3-70b-versatilemodel to process conversational queries in milliseconds, avoiding the slow loading times of traditional AI APIs. - Universal App Launcher: Integrated with
AppOpener, Jarvis can search your entire Windows system and instantly launch any installed application (e.g., "Open Spotify", "Open Settings", "Open Chrome") or websites. - System Control: Built-in OS commands allow you to securely manage your laptop just by speaking:
- "Lock the PC"
- "Shut down the computer"
- "Restart the laptop"
- Persistent Memory System: A local JSON-based memory manager allows you to ask Jarvis to remember specific facts (e.g., "Remember that my favorite color is blue") and recall them later ("What is my favorite color?").
- Push-to-Talk Interface: A minimalistic CustomTkinter GUI that replaces slow 24/7 background listening with a highly responsive "Click to Speak" button, guaranteeing perfect audio capture without hitting API rate limits.
- Thread-Safe TTS: Uses
pyttsx3forced onto the main Tkinter thread to completely bypass notorious Windows COM audio freezing bugs, ensuring the robotic voice always works perfectly.
The script is cleanly divided into modular classes:
MemoryManager: Handles reading/writing tomemory.json.VoiceManager: Usessounddeviceto record raw PCM audio andspeech_recognitionto convert it to text. Handles TTS viapyttsx3.AIManager: Maintains a conversational context array and connects to the Groq API.CommandProcessor: An intent-router that intercepts hardcoded system commands (like launching apps or checking the time) before falling back to the AI for general questions.JarvisApp: The main Tkinter application that ties all managers together and handles multithreading.
Requires Python 3.14+ (or lower) with the following dependencies:
pip install sounddevice numpy speechrecognition pyttsx3 customtkinter plyer groq AppOpener