Speak it once. Done.
Who and when.
Optional. The model uses these to address the patient by name.
Speak the visit. Loosely.
Subjective findings, exam, assessment, plan. Don’t worry about structure — the model will sort it.
Four steps, finished
before your coffee gets cold.
MedScribe is built for the moment right after a patient leaves the room — when memory is sharpest and time is shortest.
Start a RecordingYour patients’ data never leaves your device.
Capture
Dictate after the visit. Three minutes is plenty.
Transcribe
Whisper turns the dictation into a clean transcript.
Draft
The model writes a SOAP note and a plain-English handout.
Review
Read, edit, export. Nothing leaves until you say so.
Built around UAE health data law.
Federal Law No. 2 of 2019, the PDPL, and DHA’s AI Policy don’t ask for marketing — they ask for architecture. Here is ours.
Patient data stays in the UAE.
Federal Law No. 2 of 2019 keeps UAE health data inside the country. Our backend runs only on UAE-hosted infrastructure — no cross-border transfer.
The model runs on your device.
SOAP notes are drafted on the clinician’s own machine. No PHI ever reaches OpenAI, Anthropic, or any third-party LLM. Your patients train no one.
Audio is discarded after the summary.
Recordings live only in the browser, used once to produce the transcript, then discarded. Nothing hits disk. There is no archive to breach.
Your EMR is still the record.
We don’t create a parallel patient database. The final note lives only in the clinic’s EMR — already on the hook for the federal 25-year retention rule.
Notes that finish themselves.
Memory is sharpest the moment after a visit ends. MedScribe turns that minute into a signed note and a plain-English handout for the patient who just walked out.