R6 Nice Talker

A python script that uses AI to generate and type "wholesome" (or edgy, or salty) messages in Rainbow Six Siege chat.

Features

• AI-Powered: Uses OpenAI (ChatGPT) to generate unique, context-aware messages.
• Persona Switching: Cycle between personas like "Reputation Farmer", "Preacher", "Salty", etc. using hotkeys.
• Game Safe: Simulates human-like keystrokes (holds keys for 50ms) to ensure input registers in-game.
• Context Aware: Simulates game events (clutches, losses) to generate relevant banter.
• Audio Feedback: Beeps to confirm actions or persona switches so you don't have to tab out.
• Voice Injection: Text-to-Speech output directly into the game's voice chat via Virtual Cable.
• TUI Mode: Optional terminal-based UI for real-time monitoring (use --tui flag).

Setup

1. Install Dependencies:

   pip install -r requirements.txt

2. Configure:

   • Copy env.example to .env.
   • Add your OPENAI_API_KEY.
   • (Optional) Customize TRIGGER_KEY (default F6) or NEXT_PROMPT_KEY (default F8).

3. Language Support:

   • By default, the bot speaks English (en).
   • To change to Portuguese, update .env:

     LANGUAGE=pt-br

4. Run:

   • Must run as Administrator to interact with the game window.

   # Console mode (default)
   python main.py
   
   # TUI mode (terminal UI with real-time monitoring)
   python main.py --tui

Voice Injection Setup

To play AI-generated voice lines directly into the game's voice chat:

1. Install a Virtual Audio Cable:

   • Download and install VB-Cable (or similar).
   • This creates a virtual input ("CABLE Input") and output ("CABLE Output").

2. Configure Game Audio:

   • In Rainbow Six Siege settings, set Voice Chat Record Mode to "Push to Talk" (or "Open" if you dare).
   • Set Voice Chat Input Device to "CABLE Output".

3. Configure Bot Output:

   • Run the device discovery tool to find your virtual cable's exact name:

     python list_devices.py

   • Update your .env file with the device name:

     AUDIO_OUTPUT_DEVICE_NAME="CABLE Input (VB-Audio Virtual Cable)"

   • Alternatively, you can use the device index (AUDIO_OUTPUT_DEVICE_INDEX), but names are more reliable.

Text-to-Speech Configuration

You can choose between the built-in offline TTS (pyttsx3) or high-quality AI voices (ElevenLabs).

Option 1: Offline (Default)

• Uses your system's installed TTS voices.
• Fast, free, but robotic.
• Configuration in .env:

  TTS_PROVIDER=pyttsx3

Option 2: ElevenLabs (High Quality)

• Uses AI generated voices.
• Requires an API key.
• Configuration in .env:

  TTS_PROVIDER=elevenlabs
  ELEVENLABS_API_KEY=your_api_key
  ELEVENLABS_VOICE_ID=21m00Tcm4TlvDq8ikWAM

  • ELEVENLABS_VOICE_ID can be found in the VoiceLab on their website.

Usage (Voice)

1. Press F5 (default) to generate and speak a message into the voice chat.
2. Note: The bot plays the audio into the virtual microphone. You may need to hold your Push-to-Talk key in-game while the bot is speaking.

Usage

Running the Bot

Console Mode (Default):

python main.py

TUI Mode (Terminal UI):

python main.py --tui

The TUI provides real-time monitoring of:

• Bot status and components
• Current persona and hotkeys
• Session statistics (API calls, costs, latency)
• Event log with timestamps

Press q or Ctrl+C to exit TUI mode.

Hotkeys

• F6: Generate and type a message.
• F5: Generate and speak a message.
• F8: Next Persona.
• F7: Previous Persona.
• Ctrl+C: Quit.

Customization

Edit prompts.json to add your own personas or change the style of existing ones.

Documentation

• Development Guide - Testing tools, debugging, and cost management
• TUI Guide - Terminal User Interface features and usage
• Architecture - Component diagrams and flow documentation
• Learnings - Design patterns and lessons learned
• Roadmap - Future features and enhancements