Emosphere is an innovative AI mental health companion designed to help users proactively manage their emotional well-being through voice analysis and personalized interventions. In a world where stress and mental health challenges are increasingly prevalent, Emosphere provides an accessible and private tool to understand and respond to one's emotional state.

The application works by analyzing a user's voice patterns to detect their current stress or emotion level (calm, mild, moderate, high). This is achieved through a sophisticated backend voice stress detection system. Based on the detected emotion, Emosphere dynamically adapts its user interface and recommends tailored interventions, such as guided breathing exercises, grounding techniques, or cognitive reframing.

Key features include:

Voice Stress Detection: Real-time analysis of voice to identify emotional states.
Adaptive Interface: The UI adjusts its theme and suggestions based on the user's detected emotion level.
Personalized Interventions: Access to a library of evidence-based techniques (breathing, grounding, cognitive, mindfulness) recommended specifically for the user's current state.
Session History & Mood Insights: Track emotional trends over time and identify potential triggers.
Ambient Soundscapes: Curated audio environments to aid relaxation and focus.
Interactive AI Chatbot: A conversational interface for immediate support, guidance, and access to interventions.
Progressive Onboarding: Guides new users through the application's core functionalities.
Crisis Support Integration: Provides quick access to emergency resources when needed.

Emosphere aims to empower individuals with self-awareness and practical tools to navigate their emotional landscape, fostering resilience and promoting a balanced mental state.

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for EmoSphere - AI Mental Health Companion

Built With

  • axios
  • date-fns
  • elevenlabs
  • flask
  • framer
  • framer-motion
  • frontend:-react
  • librosa
  • lodash.debounce
  • lodash.throttle
  • lucide
  • lucide-react
  • nanoid
  • numpy
  • python
  • react
  • react-helmet-async
  • react-use
  • recharts.-backend-(voice-stress-detection):-python
  • scipy.-database:-supabase-(postgresql).-ai/ml:-tensorflow-lite-(for-voice-stress-detection-model)
  • soundfile
  • supabase
  • tailwind
  • tailwind-css
  • tensorflow
  • typescript
  • vite
  • vitechlab
  • zustand
Share this project:

Updates