Rose is your AI Shaman healer — an empathetic voice-first wellness companion that combines ancient healing wisdom with modern AI. She listens through natural push-to-talk voice, remembers your journey across sessions, and responds with a calming, therapeutic presence powered by Groq's ultra-fast inference.
Rose is not a chatbot. She is a persistent, voice-first AI companion designed for therapeutic conversations. She remembers your past sessions through vector-based long-term memory, speaks with a natural calming voice via ElevenLabs, and maintains an empathetic healer personality trained in ancient healing traditions.
| Feature | Description |
|---|---|
| Voice-First Interaction | Natural push-to-talk interface for hands-free, immersive conversations |
| Therapeutic Personality | Empathetic healer shaman trained in ancient healing traditions |
| Persistent Memory | Remembers your journey across sessions using Qdrant vector database |
| Groq-Powered Inference | Ultra-fast responses using Llama 3.3 70B and Whisper Large v3 |
| Natural Voice | Calming text-to-speech powered by ElevenLabs |
| Beautiful UI | React-based interface with smooth animations and visual feedback |
Python 3.12+, Node.js 18+, and the uv package manager are required. You will also need API keys for Groq, ElevenLabs, and Qdrant.
# Clone the repository
git clone https://github.com/Alexi5000/Rose.git
cd Rose
# Install Python dependencies
uv sync
# Install frontend dependencies
cd frontend && npm install && cd ..
# Configure environment variables
cp .env.example .env
# Add your API keys to .env# Start the backend
uv run python -m ai_companion.app
# In a separate terminal, start the frontend
cd frontend && npm run devVisit http://localhost:5173 to begin your session with Rose.
| Category | Technology |
|---|---|
| Orchestration | LangGraph (workflow & state management) |
| LLM Inference | Groq API (Llama 3.3 70B) |
| Speech-to-Text | Groq Whisper Large v3 |
| Text-to-Speech | ElevenLabs |
| Vector Memory | Qdrant |
| Backend | FastAPI (Python 3.12+) |
| Frontend | React + TypeScript |
| Package Manager | uv |
Rose uses LangGraph for workflow orchestration, routing voice input through Groq's Whisper for transcription, processing through Llama 3.3 70B with context from Qdrant's vector memory, and responding through ElevenLabs' natural voice synthesis. The entire pipeline runs in under 2 seconds for a seamless conversational experience.
Contributions are welcome. Please open an issue or submit a pull request. All contributions must comply with the MIT license.
This project is licensed under the MIT License. See the LICENSE file for details.
Maintained by Alex Cinovoj | TechTide AI
Originally created by neural-maze