Voicexec

Voice command system that processes natural language to create calendar events, manage todos, and more.

See it in action: Day in the Life - Building Voicexec

Features

• Voice-to-text transcription using OpenAI Whisper
• Natural language intent processing with local LLM (LM Studio)
• Google Calendar integration
• Todo list management
• Real-time audio recording in browser

Architecture

voicexec/
├── backend/          # Express.js API (TypeScript)
├── frontend/         # Next.js web client (React)
└── whisper-service/  # Python microservice for audio transcription

Prerequisites

• Node.js 18+
• Python 3.10+
• PostgreSQL
• LM Studio running locally
• Google Cloud Console project (for Calendar API)

Quick Start

1. Backend

cd backend
npm install
cp .env.example .env
# Edit .env with your database and service URLs

npm run db:generate
npm run db:migrate
npm run dev

2. Whisper Service

cd whisper-service
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt
python main.py

3. Frontend

cd frontend
npm install
npm run dev

Open http://localhost:3000 in your browser.

Configuration

Environment Variables

See backend/.env.example for required environment variables:

• DATABASE_URL - PostgreSQL connection string
• LM_STUDIO_BASE_URL - LM Studio API endpoint (default: http://127.0.0.1:1234/v1)
• WHISPER_SERVICE_URL - Whisper service endpoint (default: http://localhost:8001)

Google Calendar Setup

1. Create a project in Google Cloud Console
2. Enable the Google Calendar API
3. Create OAuth 2.0 credentials (Desktop app)
4. Download credentials.json to the backend/ directory
5. Run the backend and visit /api/auth/google/login to authorize

Development

Each service can be run independently. See individual README files:

• Backend README
• Frontend README
• Whisper Service README

License

MIT