🎙️ LoudMouth - Privacy-First Voice-to-Text

Offline, fast, and developer-friendly voice-to-text application powered by OpenAI Whisper

✨ Download Latest Release | 📖 Documentation | 💬 Issues | ☕ Support Development

---

🚀 Features

• 🔒 100% Offline - No internet required, complete privacy
• ⚡ Fast - Optimized for short audio clips
• 🎯 Developer-Focused - Technical filter converts "underscore" → "_", "dot py" → ".py"
• 🎛️ Customizable - Multiple models, hotkeys, and settings
• 🖥️ Cross-Platform - Windows, macOS, Linux support
• 🎪 Easy to Use - Simple push-to-talk interface

📸 Screenshots

🎯 Why LoudMouth?

Privacy Matters: Unlike cloud-based solutions, LoudMouth processes everything locally
Speed: No network delays, instant transcription
Accuracy: Choose from multiple Whisper models based on your needs
Developer-Friendly: Built-in technical vocabulary for programming

📦 Quick Start

Option 1: Download Pre-built Binary

1. Go to Releases
2. Download for your platform:
   • LoudMouth-Windows.exe for Windows
   • LoudMouth-Linux for Linux
   • LoudMouth-macOS for macOS
3. Run and enjoy!

Option 2: Run from Source

git clone https://github.com/Sggin1/LoudMouth.git
cd LoudMouth
pip install -r requirements.txt
python main.py

🛠️ System Requirements

• Python 3.8+ (for source installation)
• 2GB RAM minimum (4GB recommended)
• Microphone (obviously!)
• 500MB disk space (for models)

⚙️ Available Models

Model	Size	Speed	Accuracy	Best For
tiny	~150MB	~10x	Basic	Quick notes
base	~290MB	~7x	Good	General use
small	~967MB	~4x	Better	Most users
medium	~3.0GB	~2x	Very Good	High accuracy
large	~6.0GB	1x	Best	Professional use

🔧 Configuration

Hotkeys

• Default: Shift (hold to record)
• Supports: Keyboard keys, mouse buttons
• Customizable in settings

Technical Filter

Perfect for developers! Converts spoken words to code:

• "underscore" → "_"
• "dot py" → ".py"
• "open paren close paren" → "()"
• "equals equals" → "=="
• And 100+ more conversions!

🤝 Contributing

We welcome contributions! Here's how to help:

1. 🐛 Report Bugs - Use the Issues tab
2. 💡 Suggest Features - We love new ideas!
3. 🔧 Submit PRs - Check our Contributing Guide
4. 📖 Improve Docs - Documentation is always appreciated

💖 Support Development

LoudMouth is free and open source. If you find it useful, consider supporting:

• ⭐ Star this repository (helps others discover it)
• ☕ Buy me a coffee (one-time support)
• 💝 GitHub Sponsors (monthly support)
• 🐦 Share on social media (spread the word!)

📋 Roadmap

• v1.1 - better key bind system signle or combo , removed L-click from bind list
• v1.x - two parallel pipeline
• v1.x - button release routes audio to command buffer
• v1.x - async command injection " spoken = "space 5, copy 7, enter , paste Output: 'Space', 'Space', 'Space', 'Space', 'Space', 'Copy( next 7 chars)','enter', paste
• v2.x - lightweight translucent UI overlay

🔧 Building from Source

# Install build dependencies
pip install pyinstaller

# Build executable
pyinstaller --onefile --windowed --name="LoudMouth" main.py

# Find your executable in dist/

📄 License

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

🙏 Acknowledgments

• OpenAI for the incredible Whisper model
• Contributors who make this project better
• Users who provide feedback and support

---

Made with ❤️ for developers who value privacy and efficiency

Report Bug · Request Feature · Discussions