Whose Voice?

A minimal AI-powered web application that analyzes sentences and classifies them into three categories: Personal Voice, Learned Belief, and Fear-Based.

Setup

Prerequisites

• Python 3.11+
• OpenRouter API Key (for DeepSeek access)
• Docker (optional, for containerized deployment)

Local Development

1. Create and activate virtual environment:

# Create virtual environment
python3 -m venv venv

# Activate on Mac/Linux
source venv/bin/activate

# Activate on Windows
venv\Scripts\activate

2. Install dependencies:

pip install -r requirements.txt

3. Create .env file:

cp .env.example .env

4. Add your OpenRouter API key to .env:

OPENROUTER_API_KEY=your_actual_key_here

5. Run the application:

python app.py

6. Open browser: http://localhost:8080

7. When finished, deactivate virtual environment:

deactivate

Docker

1. Make sure Docker Desktop is running

2. Build the image:

docker build -t whose-voice .

3. Run the container:

docker run -p 8080:8080 -e OPENROUTER_API_KEY=your_key_here whose-voice

4. Open browser: http://localhost:8080

5. Stop container with Ctrl+C

Usage

1. Enter a sentence describing a thought or decision
2. Click "Analyze"
3. View the breakdown across three categories
4. Read the neutral explanation

Tech Stack

• Backend: Python + Flask
• Frontend: HTML5 + Vanilla JavaScript + CSS3
• AI: DeepSeek R1 via OpenRouter API
• Deployment: Docker

Design Philosophy

This application follows these principles:

• Minimal: Single feature, zero bloat
• Neutral: No advice, no judgment
• Fast: Instant feedback, no tracking
• Private: No authentication, no data storage
• Honest: Reflects what's there without prescription

License

MIT