AI-powered radiology dictation

Dictate findings.
Get reports.

DictateMD turns spoken radiology findings into structured, professional reports — instantly. Stay in your images, not your notes.

No installation · Works in any modern browser

CT HEAD — LIVE SESSION
Raw dictation
Click "See how it works" to watch a live dictation...
Formatted report
Brain Parenchyma
No acute intracranial hemorrhage. Grey-white differentiation is preserved.
Ventricles & CSF
Ventricles are normal in size and configuration. Basal cisterns are patent. Sulci appear normal for age.
Extra-axial Spaces
No extra-axial collections. No midline shift.
Calvarium
Osseous structures are intact.
Impression
Negative head CT. No acute intracranial findings.
Works with
🌐
Chrome Speech
Deepgram
🤖
Whisper
🌊
Wispr Flow
🏥
AWS Medical
Everything a radiologist needs

Built around the real radiology workflow — from first word to final report.

🎙

Real-time transcription

Speak naturally and watch your words appear instantly. Interim results lock in as you talk. Edit anything in the transcript before processing — no restrictions, no locked fields.

"No acute intracranial hemorrhage identified, ventricles are normal size... uh, grey-white differentiation preserved, no midline shift..."

AI report formatting

Claude AI strips filler words, corrects grammar, applies proper radiological terminology, and maps everything to your template — producing a clean, professional report in seconds.

📋

Custom template library

Build and strictly enforce a personalized library of structured templates. Smart validation ensures clean organization across all modalities and categories.

📄

Learn from your PDFs

Drop in any existing report PDFs and DictateMD reverse-engineers the structure into a reusable template — automatically, in one click.

📱

Phone as a microphone

No expensive hardware needed. Securely pair your smartphone via a quick QR code to use it as a high-quality clinical microphone instantly.

⚠️

Missing field detection

Required sections not covered in your dictation are flagged before you finalize. Every acknowledgment is logged for audit purposes.

🧠

AI learning memory

Correct the AI once and it remembers. Adaptations inject into every future report until you mark them resolved.

☁️

Cloud sync & security

All templates, preferences, macros, and secure dictation logs automatically persist across all your devices using robust backend authentication.

📚

Full session audit trail

Every processed report is auto-saved — transcript, audio recording, formatted output, and any missing fields noted at the time.

🖥️

Designed for dual-screen workflows

The output panel is fully draggable — position it beside your PACS viewer and never take your eyes off the images. Minimize it when you don't need it. Copy the report with one click.

Voice providers
Works with your voice stack

Start free with Chrome's built-in recognition, or connect Deepgram's medical model for best-in-class clinical vocabulary accuracy.

🌐 Web Speech API Live
Deepgram Nova-2 Medical Live
🤖 OpenAI Whisper Live
🌊 Wispr Flow Live
🔊 AssemblyAI Soon
🏥 AWS Transcribe Medical Soon
📋 AWS HealthScribe Soon
☁️ Azure Speech Soon
"I used to spend 40% of my time on report formatting. Now I just speak and it's done."
Dr. S. Chen · Consultant Radiologist

Start dictating smarter today

No installation. No account. Open the app and start your first dictation in under a minute.

Works in any modern browser · Secure cloud synchronization