🎙️ WisprFlow Lite

A powerful voice-to-text transcription system that works anywhere on your Mac. Hold down the Option/Alt key while speaking, and release it to have your words transcribed directly where your cursor is positioned. Powered by Fireworks AI's Whisper V3 Turbo for fast, accurate transcription.

📱 Two Versions Available

Choose the version that best fits your needs:

🍎 macOS Native App (Recommended)

• Location: wispr-flow-mac/
• For: All macOS users
• Installation: Build once with ./build_app.sh, drag to Applications
• Interface: Native menu bar app with SwiftUI settings
• Performance: ⚡ Lightning fast (native Swift)
• Status: ✅ Production ready

🖥️ Python CLI Version

• Location: python-cli/
• For: Developers, power users, and cross-platform use
• Installation: Python virtual environment
• Interface: Command-line interface
• Performance: Fast (Python + Fireworks AI)
• Status: ✅ Production ready

✨ Features

• 🎯 Push-to-Talk - Hold Option key to record, release to transcribe
• 🎤 High-quality transcription - Uses Fireworks AI Whisper V3 Turbo
• ⚡ Ultra-fast - 66% faster than OpenAI Whisper
• 📝 Direct typing - Text appears where your cursor is
• 🌍 Multi-language - Supports 100+ languages
• 🔧 Customizable - Configure language, typing speed, and more
• 🔄 Robust Error Handling - Automatic retries for API calls and better resource management
• 📊 Memory Management - Prevents crashes during long recordings
• 🔍 Advanced Logging - Better error tracking and debugging

🚀 Quick Start

macOS Native App (Recommended)

1. Navigate to the directory:

   cd wispr-flow-mac

2. Build the app:

   ./build_app.sh

3. Install:

   • Drag Wispr Flow Lite.app to your Applications folder
   • Launch the app
   • Grant Microphone and Accessibility permissions
   • Enter your Fireworks AI API key in Settings

For detailed instructions, see wispr-flow-mac/README.md.

Python CLI Version

For the command-line version, see detailed instructions in python-cli/README.md.

Quick setup:

cd python-cli
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
cp .env_example .env  # Edit with your Fireworks API key
python voice_transcriber_fireworks.py

💰 Cost & Privacy

• Fireworks AI Whisper V3 Turbo: ~$0.002 per minute (66% cheaper than OpenAI)
• No audio stored locally
• Data sent to Fireworks AI for transcription
• Monitor usage: Fireworks AI Dashboard

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Fireworks AI for the Whisper V3 Turbo API
• Original WisprFlow app for inspiration
• All contributors and users

---

Made with ❤️ by Tommy Yau