Dragon Medical One
The clinical speech standard: cloud dictation across 56+ medical specialties, rated #1 in KLAS for front-end EMR speech recognition.
About
Dragon Medical One is Nuance's cloud speech platform for clinical documentation, now under Microsoft. Clinicians dictate directly into the EHR or any application, with a vocabulary covering dozens of medical specialties and no voice-profile training: a single cloud profile adapts automatically to accent and microphone.
It is HIPAA-aligned, hosted on Microsoft Azure with HITRUST CSF certification, and rated Best in KLAS for front-end EMR speech recognition for five consecutive years. Nuance claims accuracy up to 99%, which should be read as a vendor figure for clean dictation rather than noisy wards. Pricing is an annual per-user subscription quoted through Microsoft and partners, and a successor AI assistant called Dragon Copilot is in the works.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
- Up to 99% accuracy with automatic accent detection (provider-claimed)
- Best in KLAS for Speech Recognition: Front-End EMR, 5 consecutive years (provider-reported)
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (metrics tier, metrics-v1). Evidence: Microsoft Azure hosting with HITRUST CSF certification from Nuance's own white papers; Best in KLAS front-end EMR speech recognition (5 years) and the up-to-99% accuracy claim from Nuance product materials; 56+ specialty vocabulary from reseller docs. Strengths: the entrenched clinical standard with the deepest medical vocabulary, strong compliance posture, single cloud voice profile. Docked for: (1) accuracy and KLAS figures are provider-reported; (2) closed ecosystem with opaque per-user subscription pricing and Windows-centric clients. License: proprietary (commercial license).
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 10.0/10
- Maintenance · recency, releases, issue load
- 8.0/10
- Openness · license + docs
- 1.0/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · Best in KLAS front-end EMR speech recognition 5 consecutive years; up to 99% accuracy (provider-claimed)
- · Hosted on Microsoft Azure, HITRUST CSF certified; 56+ medical specialty vocabularies
- · Successor 'Dragon Copilot' AI assistant announced