VoiceIntake v1.0.0 — Initial Production Release
VoiceIntake is an AI-powered patient intake platform that allows patients to complete medical intake forms by speaking naturally.
The system transcribes patient speech using Whisper, extracts structured patient information using Llama 3.3, and instantly generates a completed intake record for clinic staff.
Highlights
Voice-Based Patient Intake
- Patients can provide information through natural speech instead of manually filling forms.
- Reduces front-desk administrative workload and patient waiting time.
AI-Powered Data Extraction
- Automatic extraction of:
- Full name
- Date of birth
- Address
- Phone number
- Insurance information
- Symptoms
- Emergency contact details
Multi-Language Support
- English
- Arabic
- German
- French
Multi-Clinic Architecture
- Clinic-specific sessions and data isolation.
- Embeddable intake widget for multiple clinics.
Security & Compliance
- Authentication with Supabase Auth.
- Consent workflow before recording.
- Audio processed in memory and not permanently stored.
- Environment-based configuration and protected API routes.
Staff Dashboard
- Secure clinic dashboard.
- Real-time session management and review.
- Searchable patient intake records.
Tech Stack
- React + Vite
- FastAPI
- Groq Whisper
- Groq Llama 3.3 70B
- Supabase PostgreSQL
- Supabase Authentication
- Railway
- Vercel
Release Notes
This release marks the first stable production version of VoiceIntake and includes the complete patient intake workflow, staff dashboard, authentication system, multi-language support, multi-clinic support, deployment configuration, and documentation.
Built by Ahmed Al-Majid.