Skip to content

Ak123yay/ACL-HACKS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

38 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MIRROR — Talk to Yourself. Heal Yourself

Mirror is an AI-powered mental wellness application that creates a personalized AI clone of the user using their own words, voice, and thinking patterns. It allows users to talk through stress, decisions, and emotions with a version of themselves that reflects their own reasoning and values.

Overview

Many people experience stress and anxiety but lack access to immediate, affordable, and personalized support. Existing solutions are often expensive, generic, or unavailable when needed.

Mirror addresses this by enabling users to interact with an AI system built entirely from their own data, providing responses that feel familiar, personal, and relevant.

Key Features AI Persona Generation Builds a personalized AI model from user data and reflection inputs Voice Cloning Generates responses in the user’s own voice Retrieval-Augmented Memory Uses past messages to provide context-aware responses Conversation Modes Vent: listens and reflects without giving advice Reframe: helps challenge negative thinking Boost: provides motivation and confidence Session Summaries Generates short reflections after each conversation Mood Tracking Tracks emotional changes before and after sessions Crisis Detection Detects high-risk inputs and surfaces support resources Technology Stack

Frontend

React (Vite) Tailwind CSS Web Audio API

Backend

FastAPI (Python) Supabase (database and authentication)

AI Services

Claude API (persona generation and chat) OpenAI Whisper (speech-to-text) ElevenLabs (voice synthesis and cloning)

Data Layer

ChromaDB (vector database for memory retrieval) How It Works The user uploads personal data such as chat exports The user answers reflection questions The user records a short voice sample The system generates a personalized AI persona The user begins interacting with the system through voice or text Getting Started Clone the Repository git clone https://github.com/your-username/mirror.git cd mirror Frontend Setup cd frontend npm install npm run dev Backend Setup cd backend python -m venv venv

Mac/Linux

source venv/bin/activate

Windows

venv\Scripts\activate

pip install -r requirements.txt uvicorn main:app --reload Environment Variables

Create a .env file in the backend directory:

SUPABASE_URL=your_url SUPABASE_KEY=your_key ANTHROPIC_API_KEY=your_key OPENAI_API_KEY=your_key ELEVENLABS_API_KEY=your_key Project Structure mirror/ ├── frontend/ # React application ├── backend/ # FastAPI server ├── chroma_data/ # Vector database storage Hackathon Scope

Minimum viable demo should include:

Chat system with AI responses Basic persona generation Voice output

Optional features:

Insights dashboard Journal system Full authentication flow Disclaimer

Mirror is intended as a self-care tool and does not replace professional mental health services. Users in crisis should seek immediate help from qualified professionals.

Future Improvements Mobile application support Advanced analytics and insights Multi-persona support (mentor, coach, etc.) Improved personalization models

About

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors