RadMyk

RadMyk is medical dictation software for radiologists, sold as a one-time purchase instead of a subscription.

The speech recognition model runs entirely on the radiologist's own machine. No audio is streamed to a cloud server, no vendor receives voice data, and the software keeps working when the network does not. After a one-time engine download, no internet connection is required.

RadMyk types at the text cursor like a keyboard. If you can type into it, you can speak into it: RIS and PACS report fields, EHR text boxes, browser-based reporting tools, Microsoft Word, and Citrix or remote desktop sessions. There is no PACS integration project and no new workflow to learn.

The model is tuned on radiology language - anatomy, modalities, measurements, laterality, and the cadence of structured reporting. Measured word accuracy is 96.1% out of the box, transcribing in real time at roughly 220 words per minute. A guided calibration step tunes it to your voice and microphone, and voice macros expand spoken triggers into full report sections.

RadMyk runs natively on macOS (Apple Silicon) and Windows.

Pricing is one-time: $199 for the first 100 users (14,999 INR in India), future updates included, no renewals, no per-seat subscription. There is a 14-day free trial with no credit card, an extended trial for verified radiologists, and radiology trainees use RadMyk free for the entire length of their training.

What RadMyk is not: an enterprise reporting platform (no peer review or structured encoding), an ambient AI scribe, or a cloud service. It does one job - private, fast, owned dictation - and does it anywhere you can type.