FR3K Talk - Complete Voice System for Claude Code

🎤 Production-ready voice interface with auto-listen, auto-send, and voice interruption support.

🚀 Quick Start

# Install globally
npm install -g fr3ktalk

# Install voice system (auto-detects and installs all dependencies)
fr3ktalk install

# Start continuous voice listening for Claude Code
fr3ktalk listen

# Or start with Claude integration
fr3ktalk claude

✨ Features

• 🎙️ Auto-Listen - Continuous voice monitoring
• 📤 Auto-Send - Automatic transcription to Claude Code
• 🔇 Voice Interruption - Interrupt TTS by speaking
• 🧠 High-Quality Voices - EdgeTTS neural voices
• 📱 Android Optimized - Built for Termux
• ⚡ Fast Setup - One-command installation with testing
• 🛠️ Robust - Multiple fallback engines and error handling

📋 System Requirements

• Node.js 14+
• Python 3.7+
• Android/Termux (recommended) or Linux
• Microphone permission for Termux

🛠️ Installation

Automatic Installation (Recommended)

npm install -g fr3ktalk
fr3ktalk install

The installer will:

• ✅ Check system requirements
• ✅ Install Termux packages (ffmpeg, termux-api, mpv)
• ✅ Install Python packages (speechrecognition, edge-tts, pyaudio)
• ✅ Setup voice scripts with proper permissions
• ✅ Test TTS and STT functionality
• ✅ Verify complete system integration

Manual Dependencies (if auto-install fails)

# Termux packages
pkg install -y ffmpeg termux-api portaudio flac mpv

# Python packages
pip3 install speechrecognition edge-tts pyaudio

# Grant microphone permission in Android Settings
# Settings > Apps > Termux > Permissions > Microphone: ENABLED

🎤 Usage

Start Voice Integration

# Start continuous listening (default)
fr3ktalk listen

# Custom listening duration
fr3ktalk listen --duration 10

# Different voice
fr3ktalk listen --voice en-US-AriaNeural

# Disable speech output
fr3ktalk listen --no-tts

Direct Claude Integration

# Start with Claude Code integration
fr3ktalk claude

# Single command mode
fr3ktalk claude --mode single

# Continuous mode (default)
fr3ktalk claude --mode continuous

Text-to-Speech

# Simple text to speech
fr3ktalk talk "Hello, this is FR3K Talk"

# Different voice
fr3ktalk talk "Testing voice" --voice en-US-JennyNeural

System Testing

# Test all voice components
fr3ktalk test

🎯 Voice Commands

When using continuous mode, you can say:

• "stop listening" - Pause voice monitoring
• "start listening" - Resume voice monitoring
• Any other text - Send to Claude Code for processing

🔧 Configuration

Available Voices

• en-US-GuyNeural (Male, default)
• en-US-AriaNeural (Female)
• en-US-JennyNeural (Female)
• And many more EdgeTTS neural voices

Script Locations

After installation, scripts are located at:

• ~/.fr3ktalk/scripts/edgetts.py - TTS engine
• ~/.fr3ktalk/scripts/live_voice.sh - STT engine
• ~/.fr3ktalk/scripts/voice_claude.py - Claude integration

🚨 Troubleshooting

Microphone Not Working

# Install Termux:API
pkg install termux-api

# Test microphone
termux-microphone-record -f test.m4a -l 3

# Grant permission in Android Settings
# Settings > Apps > Termux > Permissions > Microphone: ENABLED

TTS Not Working

# Test EdgeTTS
python3 -c "import edge_tts; print('EdgeTTS available')"

# Test audio playback
fr3ktalk talk "Audio test"

STT Not Working

# Test speech recognition
python3 -c "import speech_recognition; print('SpeechRecognition available')"

# Test live STT
fr3ktalk test

Reinstall System

# Clean reinstall
rm -rf ~/.fr3ktalk
fr3ktalk install

📊 Performance Metrics

• TTS Latency: ~1-2 seconds
• STT Latency: ~3-5 seconds (including recording)
• CPU Usage: Minimal during idle
• Memory Usage: ~50MB for voice system
• Battery Impact: Low when not actively listening

🔍 Advanced Usage

Custom Voice Scripts

You can directly use the installed scripts:

# Direct TTS
python3 ~/.fr3ktalk/scripts/edgetts.py "Custom message"

# Direct STT
bash ~/.fr3ktalk/scripts/live_voice.sh 5

# Claude integration
python3 ~/.fr3ktalk/scripts/voice_claude.py --mode continuous

Integration with Other Applications

The voice scripts output to stdout, making them easy to integrate:

# Pipe STT output to other commands
bash ~/.fr3ktalk/scripts/live_voice.sh 5 | your_command

# Use in other applications
const transcription = execSync('bash ~/.fr3ktalk/scripts/live_voice.sh 3').toString();

🆚 FR3K Talk vs Other Solutions

| Feature | FR3K Talk | Standard Solutions | |---------|-----------|-------------------| | Auto-Listen | ✅ Built-in | ❌ Manual only | | Voice Interruption | ✅ Native support | ❌ Complex setup | | Android Optimization | ✅ Termux-native | ❌ Poor support | | One-Command Install | ✅ Full testing | ❌ Manual configuration | | Claude Integration | ✅ Ready-to-use | ❌ Custom development | | Neural Voices | ✅ EdgeTTS | ❌ Robotic voices |

🎯 Success Criteria

Your FR3K Talk installation is successful when:

• [ ] fr3ktalk install completes without errors
• [ ] fr3ktalk talk "test" speaks clearly
• [ ] fr3ktalk test shows successful STT transcription
• [ ] fr3ktalk listen starts continuous monitoring
• [ ] Voice commands ("stop listening", "start listening") work
• [ ] Microphone permissions are granted in Android

📞 Support

If you encounter issues:

1. Run diagnostics: fr3ktalk test
2. Check permissions: Ensure Termux has microphone access
3. Reinstall: rm -rf ~/.fr3ktalk && fr3ktalk install
4. Check dependencies: Verify all Termux packages are installed

🗂️ License

MIT License - see LICENSE file for details.

FR3K Talk - Your complete voice interface for Claude Code on Android!

Part of the FR3K PAI ecosystem - Android-Adapted Personal AI Infrastructure