🤖 AI Voice Assistant Jarvis for Discord

An advanced, real-time voice assistant integrated into Discord. This bot combines Speech-to-Text (STT), Large Language Models (LLM), and Text-to-Speech (TTS) to engage in natural, meaningful conversations with users in voice channels.

It features tool usage (Web Search, Time), silence detection, and dual-language support (English/Polish).

✨ Key Features

• ⚡ Real-time Transcription (STT):
  • Supports Local inference using Faster-Whisper (CUDA recommended).
  • Supports Cloud inference using Groq API (Whisper V3) for ultra-low latency.
• 🧠 Intelligent Responses (LLM):
  • Powered by Llama 3 70B via Groq API.
  • Context-aware conversations with memory management.
• 🗣️ Natural Voice (TTS):
  • High-quality voice synthesis using edge-tts (Microsoft Azure Neural voices).
• 🛠️ Autonomous Tools:
  • Web Search: Fetches real-time data (news, weather, facts) using Tavily API.
  • Time Check: Provides accurate local time and date.
• 🎙️ Smart Audio Handling:
  • VAD (Voice Activity Detection): Automatically detects silence to process speech.
  • Wake Words: Configurable trigger words (e.g., "Jarvis", "Garmin") to activate the bot (optional).

🛠️ Prerequisites

• Python 3.10+
• FFmpeg: Essential for audio processing. Must be installed and added to your system's PATH.
  • Windows: winget install ffmpeg
  • Linux: sudo apt install ffmpeg
• API Keys: You will need keys for Discord, Groq, and Tavily.

📦 Installation

1. Clone the repository:

   git clone https://github.com/Inexpli/Discord-Jarvis
   cd Discord-Jarvis

2. Install dependencies: It is recommended to use a virtual environment.

   pip install -r requirements.txt

3. Configuration (.env): Create a .env file in the root directory and add your credentials:

   BOT_TOKEN=your_discord_bot_token
   GUILD_ID=your_server_id
   GROQ_API_KEY=your_groq_api_key
   TAVILY_API_KEY=your_tavily_api_key
   HF_TOKEN=optional_huggingface_token

4. Customize Settings: Open config.py to adjust:

   • LANGUAGE: Set to "en" for English or "pl" for Polish.
   • RUN_LOCALLY: Set to True to use local GPU resources, or False to use Groq API.
   • TRIGGERS: Add or remove wake words.

🚀 Usage

1. Start the bot:

   python main.py

2. Discord Commands:
   • /join: The bot joins your current voice channel and starts listening.
   • /stop: The bot leaves the channel.
3. Interaction:
   • If REQUIRE_TRIGGER = True, start your sentence with "Jarvis" (or other configured triggers).
   • If REQUIRE_TRIGGER = False, the bot will respond to all speech detected.
   • The bot will listen, process your request, and reply via voice.

⚙️ Project Structure

• main.py: Core logic, Discord event handling, audio processing pipeline, and LLM integration.
• config.py: Configuration parameters, prompt templates, and environment variable loading.
• .env: storage for sensitive API keys (excluded from version control).

📋 Requirements (requirements.txt)

Ensure your requirements.txt includes the following libraries:

py-cord
faster-whisper
python-dotenv
groq
tavily-python
edge-tts
numpy

📄 License

This project is licensed under the MIT License.

⚠️ Limitations

Single Instance: The bot currently utilizes global variables for conversation state. It is designed to work on one server/channel at a time. Running it on multiple servers simultaneously may cause conversation history overlap.

Local Performance: If RUN_LOCALLY = True, a decent GPU (NVIDIA) is required for Faster-Whisper to run smoothly.