Inspiration
Mental health challenges affect millions of students worldwide, yet many avoid seeking help because of stigma, lack of access, high costs, or fear of judgment.
We wanted to create a solution that feels more human, more empathetic, and more accessible. This inspired us to build zeo.ai, an emotion-aware AI mental health platform that combines conversational AI, facial and vocal emotion recognition, and intelligent avatars to provide personalized support anytime, anywhere.
What it does
zeo.ai is an AI-powered digital mental health platform designed for students and educational institutions.
Key Features
- Emotion-aware AI companion with facial and vocal emotion recognition
- Real-time AI video avatar that communicates naturally with users
- Anonymous guest support without mandatory authentication
- AI chat assistant for emotional guidance and everyday conversations
- University-integrated counselor booking system
- Student and counselor dashboards
- Community platform for peer interaction and support
- Mood tracking, journaling, habit building, and wellness tools
- Multilingual support
- Educational resources, guides, videos, and relaxation content
The platform helps students receive immediate emotional support while also connecting them with professional counselors when needed.
How we built it
Frontend
- React.js
- TypeScript
- Vite
- Tailwind CSS
- Framer Motion
Backend
- Node.js
- Express.js
AI & Integrations
- Tavus AI for interactive video avatars
- Speech-to-Text (STT)
- Large Language Models (LLMs)
- Facial Emotion Recognition
- Voice Emotion Analysis
- Context Memory System
Database & Infrastructure
- MongoDB
- REST APIs
- Secure Authentication
- Role-Based Access Control
Challenges we ran into
- Building emotionally intelligent conversations that feel natural and supportive.
- Integrating multiple AI services into a seamless experience.
- Managing real-time avatar communication with minimal latency.
- Maintaining user privacy while providing personalized support.
- Creating a scalable architecture suitable for educational institutions.
Accomplishments that we're proud of
- Built a working emotion-aware AI mental health platform.
- Successfully integrated AI avatars with conversational intelligence.
- Developed a complete ecosystem including AI support, counselor booking, wellness tools, and community features.
- Enabled anonymous access to encourage students to seek help without fear of judgment.
- Created a solution aligned with UN Sustainable Development Goal 3 (Good Health & Well-being).
What we learned
Through this project we learned:
- The importance of emotional intelligence in AI systems.
- How multimodal AI can significantly improve human-computer interaction.
- The challenges of developing privacy-first mental health solutions.
- Best practices for integrating AI, speech processing, and avatar technologies.
- How to design scalable systems for healthcare and education environments.
What's next for zeo.ai
Our roadmap includes:
- Advanced emotion detection models
- Crisis intervention and emergency escalation support
- Mobile applications for Android and iOS
- Government and healthcare helpline integration
- Predictive mental health analytics
- Enhanced multilingual support
- University-wide deployment programs
- Personalized long-term wellness plans
- HIPAA and GDPR compliance improvements
Our vision is to make mental health support accessible, affordable, and available 24/7 so that no student has to face their challenges alone.
Built With
- api
- assemblyai
- azure
- css
- eleven
- express.js
- face
- framer
- groq
- labs
- mongodb
- motion
- node.js
- postgresql
- react
- services
- speech
- tailwind
- tavus
- typescript
- vite
- web
- webrtc


Log in or sign up for Devpost to join the conversation.