A fully self-hosted audio transcription and transformation app. Complete privacy, zero cloud dependencies.

Transform any Whisper app into a local, privacy-first solution

• 🔒 Complete Privacy: All audio processing happens locally - your data never leaves your machine
• 💰 Zero Cloud Costs: No API fees, no subscription costs, no usage limits
• 🚀 Offline Ready: Works without internet connection once set up
• ⚡ High Performance: Local AI models with optimized performance
• 🛠️ Fully Customizable: Use your own AI models and configurations
• 🐳 Docker Ready: One-command deployment with Docker

• Local AI: Ollama + Whisper models (replaces Together.ai)
• Authentication: NextAuth.js (replaces Clerk)
• Database: Local PostgreSQL (replaces Neon)
• Storage: MinIO or local filesystem (replaces AWS S3)
• Cache: Local Redis (replaces Upstash)
• Framework: Next.js 15 + React 19 + TypeScript
• Deployment: Docker Compose with full orchestration

1. Register/Login with local NextAuth.js authentication
2. Upload Audio to local MinIO or filesystem storage
3. AI Transcription using local Whisper models (faster-whisper)
4. Content Transformation with local LLMs (Llama 3.1, Qwen 2.5)
5. Manage History in your private local database

git clone https://github.com/skyconnfig/LocalWhisper.git
cd LocalWhisper
./deploy.sh

./quick-start.sh

# Install dependencies
npm install
# Start local services
docker-compose up -d
# Run database migrations
npm run db:push
# Start the app
npm run dev

• OS: Windows 11, macOS, or Linux
• RAM: 8GB minimum, 16GB recommended
• Storage: 10GB for AI models + your audio files
• Docker: Required for easy deployment
• Optional: GPU for faster AI processing

• base - Fast, good quality (142MB)
• medium - Recommended balance (769MB)
• large-v3 - Best accuracy (1.55GB)

• llama3.1:8b - English content processing
• qwen2.5:7b - Chinese content processing
• gemma2:9b - Lightweight alternative

After deployment:

• Main App: http://localhost:3000
• MinIO Console: http://localhost:9001 (admin UI)
• Database Studio: Run npm run studio

• 📖 Complete Deployment Guide
• ⚡ Quick Start Tutorial
• 🔧 Configuration Options
• 🏗️ Architecture Overview
• ❓ FAQ & Troubleshooting
• 📋 Full Documentation Index

Run comprehensive tests to verify your deployment:

npm run test # Full test suite
npm run test:quick # Quick functionality check
npm run test:health # System health check

LocalWhisper provides seamless migration from cloud-based Whisper apps:

• From Together.ai: Automatic local AI integration
• From Clerk: Smooth NextAuth.js transition
• From AWS S3: Local file system or MinIO setup
• From Upstash: Local Redis configuration

See Migration Guide for details.

We welcome contributions! See Contributing Guide for details.

MIT License - see LICENSE file for details.

• Built upon the excellent Whisper App foundation
• Powered by OpenAI Whisper and Ollama
• UI components from Radix UI and Tailwind CSS

⭐ Star this repo if LocalWhisper helps you build privacy-first AI applications!