Frequently Asked Questions
Find answers to the most common questions about MedConverse AI's medical transcription platform.
MedConverse AI is an AI-powered medical transcription platform that records doctor-patient consultations, identifies individual speakers in real time (speaker diarization), and generates clinical summaries automatically. It works in the browser — no special hardware required — and integrates directly with Hospital Management Systems.
When you click "Record," MedConverse AI streams audio via WebSocket to our transcription engine (powered by Soniox). You see a live transcript appear as the conversation happens, with each speaker labeled automatically. The system supports multiple languages with automatic transliteration and English translation, so consultations in any supported language produce clean, readable notes.
Speaker diarization detects "who spoke when" — automatically segmenting the transcript by speaker. Speaker identification goes a step further: by registering a ~10 second voice sample, MedConverse AI uses ECAPA-TDNN voice recognition to match speakers across sessions. So "Speaker 1" becomes "Dr. Smith" automatically, every time they speak.
MedConverse AI has built-in offline resilience. If your connection drops mid-consultation, the system continues recording audio locally using IndexedDB. When the connection is restored, the audio is automatically uploaded and processed. You'll never lose a recording due to network issues.
We provide API-key authenticated endpoints and an embeddable recording widget for Hospital Management Systems. Your HMS registers patients via API, generates a short-lived widget token, and embeds the recorder in an iframe. Session results — transcripts, AI analysis, and audio — are available via API or webhook callbacks. Set up takes minutes, not weeks.
Absolutely. MedConverse AI uses enterprise-grade security with Supabase JWT authentication, role-based access control (5-level RBAC), and Azure Blob Storage with encrypted connections. Multi-tenancy ensures complete data isolation between organizations. Audio files and transcripts are accessible only to authorized users within your organization.
After a session ends, our AI (powered by Groq) extracts a structured clinical conversation (doctor questions and patient responses), identifies key discussion points, medications, and diagnoses, and generates a concise clinical summary. Doctors can review, edit, and approve the summary before it's finalized — ensuring accuracy and clinical relevance.
We offer four tiers — Starter, Clinic, Hospital, and Enterprise — designed for practices of every size, from solo doctors to multi-location hospital networks. Each plan includes real-time transcription, speaker diarization, and AI analysis. We offer a 14-day free trial with no credit card required. Contact us for a custom Enterprise quote.