🎙️ AI Voice Customer Service Agent

An AI-powered voice assistant that enables users to call and interact with a customer service agent in real time. The assistant answers account-related questions, follows company guidelines via RAG (Retrieval-Augmented Generation), and provides fast, natural conversations using state-of-the-art speech and language models.

✨ Features

• Voice Interaction: Real-time speech-to-text and text-to-speech using Deepgram APIs.
• AI-Powered Conversations: Uses LLaMA3 model via Groq API for fast and intelligent responses.
• RAG Implementation: Ensures accurate, compliant answers by referencing company guidelines and user information.
• Python-Only Implementation: Entire pipeline (speech, RAG, AI responses) built in Python — no frontend required.
• Scalable Design: Modular architecture to extend to more services or integrate with enterprise systems.

🛠 Tech Stack

• Core Language: Python
• AI Backend: Groq API + LLaMA3
• Speech Processing: Deepgram Speech-to-Text & Text-to-Speech APIs
• Knowledge Integration: Retrieval-Augmented Generation (RAG)

🚀 Setup & Installation

1. Clone the repository

git clone https://github.com/YoshaM09/AIVoiceAssistant.git
cd AIVoiceAssistant

2. Create a virtual environment

python -m venv venv
source venv/bin/activate  # On Windows use venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Configure environment variables

• Create a .env file in the project root with your API keys:

GROQ_API_KEY=your_groq_api_key
DEEPGRAM_API_KEY=your_deepgram_api_key

5. Run the application

python demo.py

🎬 Usage

• Start the Python script

   python demo.py

• Speak into your microphone — your voice will be transcribed via Deepgram STT.
• The AI assistant (LLaMA3 via Groq API + RAG) processes your query and generates a response.
• The response is converted back to speech using Deepgram TTS and played in real time.

🤝 Contributing

• Contributions are welcome! Please submit a pull request or open an issue for suggestions.

📄 License

• This project is licensed under the MIT License.