{"id":2449,"date":"2026-06-12T10:48:24","date_gmt":"2026-06-12T10:48:24","guid":{"rendered":"https:\/\/hackmybox.com\/?p=2449"},"modified":"2026-06-12T15:47:03","modified_gmt":"2026-06-12T15:47:03","slug":"jarvis-ai-voice-assistant-project-documentation","status":"publish","type":"post","link":"https:\/\/hackmybox.com\/index.php\/2026\/06\/12\/jarvis-ai-voice-assistant-project-documentation\/","title":{"rendered":"Jarvis AI Voice Assistant Project"},"content":{"rendered":"

1. Project Objective<\/strong><\/span><\/p>

Build a fully local AI voice assistant that can:<\/p>

  • Listen through a microphone<\/li>
  • Convert speech to text using Whisper<\/li>
  • Process requests using a local LLM (Gemma4 via Ollama)<\/li>
  • Respond using text-to-speech<\/li>
  • Run continuously in a loop<\/li><\/ul>

    Goal :  to create a privacy-focused voice assistant that runs entirely on local hardware without relying on cloud AI services.<\/p><\/div><\/div>

    Stage<\/th>Component<\/th>Function<\/th><\/tr><\/thead>
    Input<\/td>Microphone<\/td>Captures user voice input in real time<\/td><\/tr>
     Speech-to-Text<\/td>Faster-Whisper<\/td>Converts spoken audio into text<\/td><\/tr>
     AI Processing<\/td>Ollama + Gemma4<\/td>Generates intelligent responses from user input<\/td><\/tr>
    Text-to-Speech<\/td>pyttsx3<\/td>Converts AI response into spoken audio<\/td><\/tr>
    Output<\/td>User<\/td>Receives spoken response from Jarvis<\/td><\/tr><\/tbody><\/table><\/div><\/div>
    <\/div><\/div>

    2. Environment Setup<\/strong><\/span><\/p>

    Create Project Directory
    mkdir ~\/jarvis
    cd ~\/jarvis
    Create Python Virtual Environment
    python3 -m venv venv
    source venv\/bin\/activate<\/p>

    Verify virtual environment:<\/p>

    which python3
    which pip<\/p>

    Expected output should point to:<\/p>

    ~\/jarvis\/venv\/bin\/<\/p><\/div><\/div>

    Step<\/th>Command<\/th>Purpose<\/th><\/tr><\/thead>
    Create Project Directory<\/td>mkdir ~\/jarvis<\/code><\/td>Creates the main project folder<\/td><\/tr>
    Move into Directory<\/td>cd ~\/jarvis<\/code><\/td>Navigates into the project folder<\/td><\/tr>
    Create Virtual Environment<\/td>python3 -m venv venv<\/code><\/td>Creates an isolated Python environment<\/td><\/tr>
    Activate Environment<\/td>source venv\/bin\/activate<\/code><\/td>Activates the virtual environment<\/td><\/tr><\/tbody><\/table><\/div><\/div>

    3. Python Dependencies<\/span><\/p>

    Install required packages:<\/p>

    pip install faster-whisper
    pip install ollama
    pip install pyttsx3
    pip install sounddevice
    pip install scipy<\/p>

    Or install all together:<\/p>

    pip install faster-whisper ollama pyttsx3 sounddevice scipy<\/p><\/div><\/div>

    Step<\/th>Command<\/th>Purpose<\/th><\/tr><\/thead>
    Install Faster-Whisper<\/td>pip install faster-whisper<\/code><\/td>Speech-to-text (Whisper model)<\/td><\/tr>
    Install Ollama<\/td>pip install ollama<\/code><\/td>Connects Python to local LLM (Gemma4)<\/td><\/tr>
    Install pyttsx3<\/td>pip install pyttsx3<\/code><\/td>Text-to-Speech engine (offline voice output)<\/td><\/tr>
    Install sounddevice<\/td>pip install sounddevice<\/code><\/td>Captures audio from microphone<\/td><\/tr>
    Install SciPy<\/td>pip install scipy<\/code><\/td>Saves audio as WAV files<\/td><\/tr><\/tbody><\/table><\/div><\/div>
    \"\"<\/div><\/div>