*A Baymax-inspired AI assistant that listens, cares, and supports.
SevaAI is a voice-enabled, emotion-aware health assistant built to offer first-level symptom support, detect emotional well-being, and act as a compassionate digital companion — especially for the elderly, chronically ill, and isolated individuals. Inspired by Baymax from Big Hero 6, SevaAI brings empathy and functionality into a practical, real-world AI prototype.
- 🎤 Voice Input & Output (Hindi/English)
- 💬 Understands user-reported symptoms using NLP
- 😟 Detects emotional tone in real-time conversations
- 🆘 Responds to emergency phrases with alerts
- 📋 Offers general well-being advice (rule-based)
- 🌐 Lightweight, offline-first design with potential for app/web integration
- Elderly home care or companionship
- First-response assistant in rural areas
- Mental health support and emotional tracking
- Health information tool for underserved communities
- Voice-based interface for non-tech-savvy users
| Component | Technology |
|---|---|
| Voice Recognition | SpeechRecognition (Google API) |
| Text-to-Speech | pyttsx3 (local TTS) |
| Emotion Detection | HuggingFace Transformers (RoBERTa) |
| Symptom Checker | Rule-based logic (can be upgraded to NLP/ML) |
| Language Support | Multilingual (Hindi/English) |
- User speaks their symptoms or feelings.
- SevaAI transcribes the speech to text using Google Speech API.
- It detects emotions using a pre-trained model (e.g., "sad", "angry", "happy").
- The assistant analyzes keywords to detect known symptoms (e.g., fever, headache).
- If urgent terms like "emergency", "pain", or "help" are detected, it responds accordingly.
- SevaAI speaks back a helpful or empathetic response in natural voice.
SevaAI is not a substitute for professional medical advice.
Always consult a licensed physician or healthcare provider. This tool is for informational purposes only. In case of a medical emergency, call emergency services immediately.
- 📱 Android/iOS integration with Flutter or React Native
- 🧠 ML-based dynamic symptom analysis (beyond rules)
- 👨⚕️ Doctor/hospital API integration (e.g., telemedicine)
- 🗂️ Session history and care reports
- 🏥 Integration with health records (FHIR standard)
- (Richa Rana and Diya Patel) – Concept, design, coding.
Pull requests and feedback are welcome. If you'd like to contribute to SevaAI, feel free to fork the repository and propose new features!
This project is open-source under the MIT License.
This project is developed for the Maverick Effect AI Challenge 2025, under the theme:
"Inclusive Healthcare Access."