Vision Voice Assistant

A macOS menu bar application that combines screenshot analysis with voice interaction using Gemini and OpenAI APIs.

Features

• Take screenshots and ask questions about them using voice or text
• Voice input using OpenAI Whisper
• Voice output using OpenAI TTS
• Screenshot analysis using Google's Gemini AI
• Easy access to screenshots folder

Setup

1. Clone this repository
2. Create a virtual environment:

   python -m venv .venv
   source .venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Copy .env.example to .env and add your API keys:

   GEMINI_API_KEY=your_gemini_api_key
   OPENAI_API_KEY=your_openai_api_key
   

Usage

Run the application:

python src/main.py

The app will appear in your menu bar with the following options:

• Take Screenshot & Ask (Voice)
• Take Screenshot & Ask (Type)
• Open Screenshots Folder

Requirements

• macOS
• Python 3.8+
• Gemini API key
• OpenAI API key

Project Structure

visionVoice/
├── src/
│   ├── __init__.py
│   ├── main.py
│   ├── app.py
│   ├── audio/
│   │   ├── __init__.py
│   │   ├── recorder.py
│   │   └── player.py
│   ├── ai/
│   │   ├── __init__.py
│   │   ├── gemini_client.py
│   │   └── openai_client.py
│   └── utils/
│       ├── __init__.py
│       └── config.py
├── config/
│   └── prompts.py
├── requirements.txt
├── .env.example
└── README.md