VoiceByte is an AI-powered multilingual hospital intake system designed to bridge communication gaps for patients who face language barriers, illiteracy, or stress when visiting hospitals. The system uses voice recognition, natural language processing, and AI-powered routing to automate patient intake and suggest appropriate medical departments.
Mission: "By giving every patient a voice, we enable faster, safer, and more inclusive healthcare."
- Language Barriers: Patients visiting hospitals in unfamiliar cities struggle to communicate symptoms in local languages.
- Manual Forms: Traditional paper forms are time-consuming, error-prone, and challenging for illiterate patients.
- Miscommunication: Errors in initial intake lead to wrong department routing and delayed treatment.
- Emergency Handling: Critical time lost in manual data collection during emergencies.
- Supports 5 Indian languages: English, Hindi, Telugu, Kannada, Malayalam
- Patients can speak naturally in their preferred language
- Real-time voice-to-text transcription
- Extracts symptoms, duration, and severity from patient speech
- Maps symptoms to appropriate medical departments
- Suggests urgency level (HIGH/MEDIUM/LOW)
- Automatically suggests the most relevant medical department
- Displays available doctors with their schedules
- Generates structured digital intake reports
- Replaces paper forms with voice-driven digital intake
- Generates printable/PDF hospital receipts
- Provides JSON output for hospital system integration
| Technology | Purpose |
|---|---|
| HTML/CSS/JavaScript | Frontend interface and interaction |
| Web Speech API | Voice recognition and text-to-speech |
| Local Storage | Patient data management |
| CSS Animations | Smooth UI/UX transitions |
| Responsive Design | Works on desktop, tablet, and mobile |
voicebyte-hospital-intake/
├── index.html # Main application interface
├── style.css # Styling and animations
├── script.js # Main application logic
├── voice-recognition.js # Voice recognition module
├── data.js # Mock hospital data and translations
└── README.md # This file
- Open
index.htmlin any modern browser (Chrome recommended) - No installation or server required
1. Language Selection → Choose from 5 Indian languages
2. Voice Interaction → Speak responses to questions
3. Symptom Analysis → AI processes and categorizes symptoms
4. Department Routing → System suggests appropriate department
5. Digital Receipt → Generate and print/download intake report
- View structured patient information
- See suggested department and doctors
- Print or save digital receipts
- Access JSON data for system integration
- Animated hospital icon with pulse effect
- Overview of key features
- Start button to begin intake
- Visual cards for 5 languages with flags
- Native language names and prompts
- Smooth transitions
- Progress bar showing intake steps
- Animated microphone button with listening indicators
- Real-time transcription display
- Assistant avatar with speech bubbles
- Professional hospital-style layout
- Patient details and symptom summary
- Urgency level badge (color-coded)
- Suggested department and doctors
- Printable/downloadable format
- JSON output for developers
- Uses Web Speech API for speech-to-text
- Supports multiple Indian languages
- Real-time interim and final transcripts
- Error handling and fallback options
- Converts assistant questions to speech
- Language-specific voice selection
- Configurable speed and pitch
const symptomMapping = {
'chest pain': 'Cardiology',
'breathing': 'Emergency Medicine',
'fever': 'General Medicine',
'bone': 'Orthopedics',
'child': 'Pediatrics'
// ... and more
};- HIGH: Emergency signs, chest pain, breathing difficulty
- MEDIUM: Fever, pain, vomiting
- LOW: Routine symptoms, general consultation
- Emergency Medicine, Cardiology, General Medicine, ENT, Orthopedics, Pediatrics, Dermatology
- Each department has doctors with availability schedules
{
"language": "hindi",
"patient": {
"name": "रमेश कुमार",
"age": "45",
"gender": "पुरुष"
},
"complaint": {
"symptoms": "सीने में दर्द और सांस लेने में तकलीफ",
"duration": "2 घंटे"
},
"triage": {
"urgency": "HIGH",
"suggested_department": "Cardiology"
}
}- Mobile App – Android/iOS apps for remote intake
- Hospital Kiosk – Voice-enabled kiosks for walk-in patients
- Ambulance Integration – Paramedic data input en route
- EHR Integration – Direct sync with hospital records
- More Languages – Additional regional and international languages
- AI Symptom Prioritization – Emergency case flagging
- Backend integration with real hospital databases
- Machine learning for improved symptom recognition
- Offline voice recognition capability
- Multi-user concurrent intake support
- ✅ Easy communication in native language
- ✅ Reduced stress and anxiety
- ✅ Faster intake in emergencies
- ✅ Better understanding of next steps
- ✅ Reduced staff workload
- ✅ Fewer intake errors
- ✅ Efficient department routing
- ✅ Digital record keeping
- ✅ Improved patient satisfaction
- ✅ Bridging language divides
- ✅ Supporting illiterate patients
- ✅ Standardizing emergency intake
- ✅ Data-driven hospital management
- Chrome/Edge recommended for best voice recognition
- Requires microphone permission
- Modern browser with Web Speech API support
- Demo uses mock hospital data
- Voice recognition accuracy varies by language/accent
- Currently frontend-only (no backend/database)
This project is licensed under the MIT License - see the LICENSE file for details.