
Dragon Medical One (DMO) is Nuance's cloud-based, AI-powered medical speech recognition platform, purpose-built for clinical documentation. Unlike consumer speech recognition (Siri, Alexa, or even Dragon NaturallySpeaking for general use), DMO is trained specifically on medical language — millions of clinical notes, specialty-specific vocabularies, medication names, anatomical terminology, and the shorthand that physicians use in daily practice. It achieves 99% accuracy out of the box for medical dictation without requiring individual voice profile training — a breakthrough that eliminated the 10-15 minute enrollment process that earlier Dragon versions required. Nuance was acquired by Microsoft in 2022 for $19.7 billion, and DMO now benefits from Microsoft's Azure cloud infrastructure and AI research investments.
As of 2026, Dragon Medical One is used by over 550,000 clinicians across 10,000+ healthcare organizations worldwide. It integrates directly with all major EHR systems — Epic, Cerner (Oracle Health), Meditech, Allscripts, athenahealth, eClinicalWorks, and others — using virtualized cursor control that allows dictation into any text field, dropdown menu, or navigation element within the EHR. The platform is HIPAA-compliant, HITRUST CSF certified, and SOC 2 Type II audited. It runs on Windows, Mac, iOS, and Android, with voice profiles that roam across devices — a physician can dictate on their office workstation, continue on an iPad during rounds, and finish on their phone at home, with the same voice profile and specialty vocabulary following them seamlessly.
DMO achieves medical dictation accuracy rates of 99% according to Nuance's published benchmarks, without requiring the 10-15 minute voice profile enrollment that earlier Dragon versions mandated. This is enabled by deep learning acoustic models trained on millions of clinical dictation hours across every medical specialty. The system includes specialty-specific vocabularies for over 90 medical specialties and subspecialties — cardiology, oncology, orthopedics, neurology, radiology, psychiatry, emergency medicine, and more — each with tens of thousands of specialty-specific terms, drug names, procedure codes, and abbreviations. DMO also adapts to individual clinician speech patterns over time, learning how a specific physician pronounces unusual drug names, patient names, or institution-specific acronyms.
DMO works with any EHR system through its virtualized cursor technology. Rather than requiring custom integration with each EHR, DMO places the dictated text wherever the cursor is active — in a note field, a dropdown, a medication order entry screen, or a problem list. This "type anywhere" approach means DMO works with Epic, Cerner/Oracle Health, Meditech, Allscripts, athenahealth, eClinicalWorks, Medhost, CPSI, and virtually any other Windows-based EHR. The system understands EHR navigation commands: "click medication orders," "select amoxicillin," "choose 500 mg," "press tab," "dictate take one capsule twice daily for 10 days" — all without touching the keyboard or mouse. This deep EHR integration is what separates medical-grade dictation from general speech-to-text tools.
A clinician's voice profile, specialty vocabulary, custom macros, and auto-text shortcuts follow them across all devices. Dictate on a Windows workstation in the office, pick up an iPad during rounds, dictate from an iPhone between patients, or use a Mac at home — DMO's cloud-based architecture ensures the same accuracy and personalized vocabulary on every device. This is critical for modern clinical workflows where physicians move between devices throughout the day. The DMO mobile app (Dragon Medical One Mobile) allows physicians to dictate notes, review schedules, and navigate the EHR from their phone or tablet. Voice commands are processed in the Microsoft Azure cloud with end-to-end encryption, meeting HIPAA and GDPR requirements for protected health information.
DMO includes a powerful macro system that allows clinicians to create custom voice commands that expand into full blocks of text or trigger multi-step actions. Common uses: a physician dictates "normal physical exam" and DMO inserts a complete, pre-written normal physical exam template: "General: Well-appearing, no acute distress. HEENT: Normocephalic, atraumatic. Pupils equal, round, reactive to light and accommodation..." — saving 2-3 minutes of dictation per patient encounter. Macros can also navigate the EHR: "pull up last echo" can open the patient's most recent echocardiogram results. Clinicians can customize macros individually or share them across departments, creating standardized documentation templates for common clinical scenarios while still allowing personalization.
DMO recognizes the full FDA medication database, including brand names, generics, dosages, and routes of administration. Dictate "start lisinopril 10 mg once daily" and DMO accurately transcribes the drug name, dosage, and frequency. The system also recognizes ICD-10 diagnosis codes — dictate "assign diagnosis code for essential hypertension" and DMO can insert the appropriate ICD-10 code (I10) into the appropriate EHR field. This feature reduces the time physicians spend searching for billing codes and decreases coding errors. DMO's medication recognition is updated quarterly to reflect new drug approvals, withdrawals, and labeling changes — critical for patient safety and accurate clinical documentation.
As a cloud-based platform handling protected health information (PHI), DMO's security architecture is audited to: HIPAA compliance, HITRUST CSF certification, SOC 2 Type II, ISO 27001, and GDPR for EU deployments. Voice data is encrypted in transit (TLS 1.3) and at rest (AES-256). DMO can be deployed within a healthcare organization's Azure tenant for data residency control. Nuance maintains business associate agreements (BAA) with covered entities. The platform supports single sign-on (SSO) via SAML 2.0, Active Directory, and Azure AD integration, with multi-factor authentication (MFA) options. Role-based access control allows administrators to configure dictation, macro, and data access permissions by provider type. Audit logging tracks all access to dictation data for compliance reporting.
Before seeing patients, open DMO on your preferred device. Your voice profile loads automatically from the cloud — no setup needed. Verify that your specialty vocabulary is active (e.g., cardiology) so DMO recognizes the specific terminology you will use. Open your EHR and navigate to the first patient's chart. Test your microphone briefly with a quick phrase to confirm audio quality — DMO works with the built-in device microphone, but many physicians use a dedicated dictation microphone (PowerMic Mobile app on their phone, or a physical dictation mic like the Nuance PowerMic III) for best accuracy in noisy clinical environments. DMO's mobile app (on iPhone or Android) can serve as a wireless microphone that transmits audio to the cloud for processing while the text appears on your workstation screen — ideal for physicians who prefer to dictate while walking or standing.
During or immediately after the patient visit, open the patient's note in your EHR. Place your cursor in the History of Present Illness (HPI) field. Dictate naturally — at your normal speaking pace, up to 160 words per minute — describing the patient's symptoms, onset, duration, severity, aggravating and relieving factors, and relevant context. DMO transcribes your words in real time with 99% accuracy. Example dictation: "The patient is a 62-year-old male with a history of hypertension and type 2 diabetes who presents with three days of substernal chest pressure radiating to the left arm, worse with exertion, relieved partially by rest, associated with mild dyspnea but no diaphoresis or nausea." DMO accurately transcribes the medical terminology, medication names, and anatomical descriptions — and gets smarter about your pronunciation patterns over successive encounters. Use voice commands to navigate: "next field" to move to the Review of Systems, "insert normal physical exam" to trigger your custom macro, "new paragraph" to start a new section.
Dictate medication orders directly into the EHR's order entry screen: "order atorvastatin 40 mg one tablet by mouth at bedtime." DMO navigates the order entry fields and enters the medication, dose, route, and frequency. For labs and imaging: "order basic metabolic panel and troponin stat this morning" and "order chest x-ray two views for today." Assign ICD-10 billing codes: "assign I25.10 for atherosclerotic heart disease" — DMO enters the code in the diagnosis field. This voice-driven ordering eliminates the need to search through dropdown menus and type medication names, which is one of the most time-consuming aspects of EHR interaction. Studies published in the Journal of the American Medical Informatics Association (JAMIA) have found that speech recognition can reduce EHR documentation time by 26-37% compared to keyboard-and-mouse entry, with the largest time savings observed in medication ordering and note dictation.
After dictating, review the note for accuracy. While DMO achieves 99% accuracy, errors still occur — particularly with unusual patient names, rare conditions, or heavily accented speech. Common error types: homophone confusion ("ileum" vs "ilium"), medication names with similar phonetics ("clonidine" vs "clonazepam"), and misinterpretation of dictated numbers or lab values. Develop a personal review routine: scan medication names and dosages first (patient safety critical), then review key clinical findings, then check formatting. Use voice commands to correct errors: "select [erroneous text]" then "correct that" to reopen the dictation window and re-dictate the phrase. Alternatively, use keyboard and mouse for corrections — DMO works alongside manual input rather than replacing it entirely. Once the note is accurate, voice-command "sign note" or click sign in the EHR. The entire documentation workflow — from starting the note to signing it — should take 3-5 minutes for a routine follow-up visit, compared to 8-12 minutes with keyboard-only entry.
A major academic medical center deployed DMO across 2,000 physicians in 2024 after a 6-month pilot with 150 clinicians in the internal medicine and emergency medicine departments. The pilot results: average documentation time per outpatient visit decreased from 8.2 minutes to 4.7 minutes (~43% reduction). Note quality scores (measured by a standardized rubric evaluating completeness, specificity, and coding accuracy) increased slightly — attributed to physicians dictating more detailed histories and assessments when they could speak rather than type. Physician satisfaction with EHR documentation improved: 72% of pilot clinicians reported they "strongly preferred" voice dictation over typing, and 68% reported spending less time on documentation at home ("pajama time"). The institution's ROI analysis showed that recovering even 15 minutes of physician time per day through faster documentation translated to significant capacity gains — effectively adding the equivalent of 12 FTE physicians' clinical time across the organization without hiring. The full deployment used Azure tenant hosting for data residency control and SSO integration with the institution's Active Directory.
An 8-physician cardiology private practice adopted DMO after their EHR transition to Epic created documentation backlogs. The practice's specific needs: cardiology terminology recognition (echocardiogram findings, catheterization reports, electrophysiology terms), integration with Epic's cardiology module, and mobile dictation for physicians who see patients across three clinic locations and two hospitals. After deploying DMO, the practice reported: average note completion time decreased from approximately 11 minutes to 5 minutes for new patient consultations and from 6 minutes to 3 minutes for follow-ups. The physicians found the cardiology-specific vocabulary particularly valuable — terms like "left ventricular ejection fraction," "paroxysmal atrial fibrillation," and medication names like "apixaban" were recognized with high accuracy from the first use. The mobile dictation feature allowed physicians to complete notes between hospital consults, eliminating the backlog of charts to complete at the end of the day. One physician in the practice reported: "I used to spend 90 minutes after clinic finishing notes at home. Now I leave when my last patient leaves."
| Plan | Pricing Model | What You Get |
|---|---|---|
| Dragon Medical One (Individual) | ~$100/clinician/month (estimated; custom quotes) | Full DMO speech recognition with specialty vocabularies, EHR integration, cross-device voice profile roaming, auto-text and macro system, medication and ICD-10 recognition, mobile app access, standard support. Minimum 1 clinician. |
| Dragon Medical One (Enterprise) | Custom annual contract (volume discounts available) | Everything in Individual plus: organization-wide deployment tools, centralized admin console, custom vocabulary and template management, SSO integration, Azure tenant hosting option, advanced analytics dashboard (dictation volume, accuracy rates, user adoption metrics), priority support, and dedicated implementation manager. Typically for 50+ clinicians. |
| DAX Copilot (add-on) | Additional cost on top of DMO | Ambient clinical intelligence: DAX Copilot listens to the entire patient-clinician conversation (with patient consent) and automatically generates a clinical note draft. Combines with DMO for editing and finalizing. This is Nuance's latest AI offering and represents the evolution from "dictation" to "ambient documentation." |
Pricing estimates verified June 2026. Nuance does not publish list pricing publicly; quotes are provided through Nuance's healthcare sales team or authorized resellers. Industry estimates place DMO at approximately $80-$120 per clinician per month for individual licenses, with enterprise agreements offering 15-30% volume discounts. DAX Copilot pricing is separate and typically ranges from $200-$400 per clinician per month depending on deployment scope. Organizations should request a formal quote based on their clinician count, EHR system, and specialty mix.
Dragon Medical One competes in the clinical documentation space alongside AI medical scribes, built-in EHR voice tools, and ambient AI solutions. Here is how it stacks up.
| Feature | Dragon Medical One (Nuance/Microsoft) | DAX Copilot (Nuance/Microsoft) | DeepScribe | EHR Built-in Dictation (e.g., Epic Voice) |
|---|---|---|---|---|
| Approach | Physician dictates directly → real-time transcription | AI listens to conversation → auto-generates note draft | AI ambient scribe: listens, summarizes, drafts note | Basic speech-to-text within EHR interface |
| Medical Accuracy | 99% (Nuance benchmark); trained on medical data | High; generates structured clinical notes from conversation | High; generates SOAP notes from conversation | Moderate; general speech-to-text not medical-optimized |
| EHR Integration | ✅ Universal — works with any Windows EHR via virtual cursor | ✅ Integrated with DMO + major EHRs | ✅ Epic, Cerner, athenahealth, eClinicalWorks | ⚠️ Works only within specific EHR; limited to note fields |
| Specialty Vocabularies | ✅ 90+ medical specialties with custom vocabularies | ✅ Inherits from DMO + ambient understanding | ✅ Learns provider-specific terminology over time | ❌ Generic vocabulary; no specialty customization |
| Voice Navigation | ✅ Full EHR navigation by voice ("click orders," "select medication") | ✅ Through DMO integration | ❌ Note generation only; no EHR navigation | ❌ Text dictation only; no navigation |
| Macros & Templates | ✅ Custom voice macros, auto-text, shared templates | ✅ Templates generated from conversation patterns | ⚠️ Limited template customization | ❌ No macro support |
| Cross-Device | ✅ Windows, Mac, iOS, Android with roaming profiles | ✅ Through DMO platform | ✅ iOS, Android, web-based | ⚠️ EHR-dependent; usually desktop only |
| Physician Effort | Must dictate note structure and content | Minimal — AI listens and drafts; physician reviews | Minimal — AI drafts; physician edits | Must dictate everything; no AI assistance |
| HIPAA Compliance | ✅ HITRUST CSF, SOC 2 Type II, BAA included | ✅ Same as DMO; patient consent required for ambient recording | ✅ HIPAA compliant, BAA included, SOC 2 Type II | ⚠️ Inherits EHR security; may not have independent certifications |
| Pricing (Estimate) | $80-$120/clinician/month | $200-$400/clinician/month (includes DMO license) | $200-$400/clinician/month | Included with EHR license; no separate cost |
| Best For | Physicians who want precise control over documentation and full EHR voice navigation | Physicians who want to minimize documentation time by letting AI draft notes from natural conversation | Organizations wanting an AI scribe separate from Nuance ecosystem with strong Epic integration | Physicians who only need basic dictation and do not want to pay for an add-on tool |
Comparison verified June 2026. The key distinction: DMO is a dictation tool — the physician controls every word. DAX Copilot and DeepScribe are ambient scribes — the AI drafts the note from conversation. Many organizations deploy DMO first and add DAX Copilot later as an upgrade, giving physicians both dictation and ambient options.
Yes. Dragon Medical One is HIPAA-compliant and Nuance (Microsoft) will sign a Business Associate Agreement (BAA) with covered entities. The platform holds HITRUST CSF certification, SOC 2 Type II attestation, and ISO 27001 certification. Voice data is encrypted in transit using TLS 1.3 and at rest using AES-256 encryption. For organizations with strict data residency requirements, DMO can be deployed within the healthcare organization's own Microsoft Azure tenant, ensuring that protected health information never leaves the organization's controlled cloud environment. Nuance's HIPAA compliance documentation is available through the Nuance Trust Center, and healthcare organizations should review these documents with their compliance and legal teams during procurement. Key operational considerations: DMO processes voice data in the cloud, not on-device, which means PHI transmitted as voice audio must be protected under the BAA. The platform's audit logging captures all access to dictation data, supporting HIPAA-required access reviews and breach investigations. Patient consent for voice recording may be required in some jurisdictions, particularly if DAX Copilot (ambient listening) is also deployed — consult your organization's legal and compliance teams for jurisdiction-specific requirements.
This is the most common comparison physicians ask about. The EHR built-in dictation tools (Epic voice recognition, Cerner speech-to-text) use general-purpose speech recognition engines — typically the same underlying technology as Windows Speech Recognition or basic cloud speech APIs. Key differences: (1) Medical vocabulary: DMO is trained on millions of hours of clinical dictation and has 90+ specialty-specific vocabularies. EHR built-in tools use generic speech recognition that frequently misrecognizes medication names, anatomical terms, and clinical abbreviations. (2) Accuracy: DMO achieves 99% accuracy for medical dictation; EHR built-in tools typically achieve 85-92% accuracy for medical content by independent estimates — meaning 8-15% more errors to correct. (3) Voice navigation: DMO allows full voice navigation of the EHR — "click orders," "select medication," "press tab" — while EHR built-in tools can only transcribe text into the active field. (4) Macros and auto-text: DMO supports custom voice macros; EHR built-in tools do not. (5) Cross-device: DMO voice profiles roam across devices; EHR built-in tools are typically device-specific. The bottom line: EHR built-in dictation is free and adequate for physicians who dictate occasionally, have low documentation volume, or work in specialties with straightforward terminology. For physicians who document heavily throughout the day and need high accuracy for complex medical terminology, DMO's accuracy and EHR navigation features typically justify the additional cost. A practical test: try dictating "The patient's transthoracic echocardiogram revealed a left ventricular ejection fraction of 35% with moderate mitral regurgitation and a small pericardial effusion" into both your EHR's built-in dictation and DMO — and compare the accuracy.
DMO works with three categories of microphones, each with different trade-offs. (1) Built-in device microphone (laptop, tablet, phone): works immediately with no additional hardware. Accuracy is adequate in quiet environments — private offices, dictation rooms — but degrades significantly in noisy clinical settings. Nuance recommends this only for light or occasional use. (2) The Nuance PowerMic Mobile app (free on iOS and Android): turns your smartphone into a wireless dictation microphone that connects to DMO on your workstation. The phone's microphone array is optimized for voice and significantly outperforms built-in laptop microphones. Most physicians using DMO in clinical settings use PowerMic Mobile as their primary microphone — it provides good accuracy in moderately noisy environments and requires no additional hardware beyond a phone. (3) Dedicated dictation microphone (Nuance PowerMic III, Philips SpeechMike): the highest accuracy option, with noise-canceling microphone arrays designed specifically for speech recognition in clinical environments. These are wired or wireless handheld microphones with programmable buttons for common DMO commands (start/stop dictation, navigate fields). Cost: $200-$400 one-time per physician. Recommended for high-volume dictation in noisy environments — emergency departments, busy hospital wards, shared offices. The PowerMic III integrates directly with DMO's command system, allowing button presses to trigger EHR navigation, not just start/stop dictation. For most physicians starting with DMO, the recommended setup is: PowerMic Mobile app for daily dictation, with an upgrade to a dedicated microphone if accuracy is insufficient in your specific clinical environment. Nuance's implementation team can assess your environment's noise levels and recommend the appropriate microphone solution.
Based on deployment data from Nuance and feedback from healthcare organizations, the typical learning curve for DMO is: Day 1-3 — basic proficiency. Learn to dictate into note fields, use the "new paragraph" and "next field" commands, and navigate within a note. Accuracy is high from the start, but dictation speed is slower than typing as the physician adjusts to speaking their thoughts rather than typing them. Week 1-2 — comfortable proficiency. Dictation speed increases to near-normal speaking pace. Physician begins using basic macros ("normal physical exam," "normal review of systems") and simple EHR navigation commands. Most physicians report matching or exceeding their typing speed by the end of week 2. Month 1-3 — advanced proficiency. Physician customizes macros, creates specialty-specific templates, and uses complex EHR navigation (ordering medications, labs, and imaging entirely by voice). Dictation speed exceeds typing speed: 140-160 words per minute spoken versus 30-50 words per minute typed for most physicians. At this stage, DMO becomes the default input method and physicians report they "cannot imagine going back to typing." The key predictor of adoption success: physicians who commit to using DMO for every note for the first two weeks — even if it feels slower initially — achieve proficiency faster than those who switch between DMO and typing. Organizations that provide 1-2 hours of initial training and designate "super users" on each clinical unit to provide peer support during the first month report significantly higher satisfaction and faster proficiency gains.
DAX Copilot (formerly Dragon Ambient eXperience) is Nuance's ambient clinical intelligence solution — an AI that listens to the entire patient-clinician conversation and automatically generates a clinical note draft. The key difference from DMO: DMO requires the physician to actively dictate the note structure and content. The physician says, "History of Present Illness: The patient is a 62-year-old male..." and DMO transcribes it. DAX Copilot eliminates the dictation step entirely. With patient consent, DAX Copilot records the natural conversation between physician and patient. Its AI analyzes the conversation, identifies clinically relevant information, and structures it into a draft SOAP note (Subjective, Objective, Assessment, Plan). The physician then reviews, edits, and signs the AI-generated draft — typically in 2-3 minutes of review time rather than 5-10 minutes of dictation. DAX Copilot is an add-on to DMO, not a replacement. Many organizations deploy DMO first and add DAX Copilot later as an upgrade, giving physicians both tools. Physicians can choose whether to use ambient mode (DAX listens and drafts) or dictation mode (DMO transcribes what they say) based on the encounter type. Nuance positions DAX Copilot as the evolution from "dictation" to "ambient documentation" — the ultimate goal being that physicians spend zero time on documentation and 100% of their attention on the patient. Pricing for DAX Copilot is separate from DMO; industry estimates place it at $200-$400 per clinician per month. Microsoft has integrated DAX Copilot into its broader Microsoft Cloud for Healthcare strategy, positioning it alongside other AI-powered clinical tools.
Yes. DMO is designed for diverse speaker populations and has been trained on speech data from clinicians with a wide range of accents, dialects, and speech patterns. Nuance reports that DMO achieves high accuracy for speakers of English with regional accents from North America, the UK, Ireland, Australia, New Zealand, South Africa, and India, as well as non-native English speakers with varying accent strengths. The system's deep learning models have been trained to recognize phonetic variations across these populations. That said, accuracy for non-native English speakers and speakers with strong regional accents may be slightly lower — Nuance reports 95-98% accuracy for these populations compared to 99% for native English speakers with neutral accents. The adaptive learning feature helps: as you correct recognition errors, DMO learns your specific pronunciation patterns and accuracy improves over the first few weeks of use. For physicians with heavy accents for whom DMO accuracy is initially insufficient, Nuance recommends: (1) Use a high-quality dedicated microphone rather than a built-in laptop mic — microphone quality disproportionately affects accuracy for accented speech. (2) Spend 15-20 minutes in the first week deliberately correcting errors so the system learns your patterns faster. (3) Use custom vocabulary entries for any words that are persistently misrecognized. (4) If accuracy remains below 95% after 2-3 weeks of consistent use and correction, contact Nuance support for accent-specific optimization — they offer specialized tuning for accent groups that are underrepresented in their base training data.
AI ambient clinical intelligence — passive note generation from natural conversations. Best for physicians who prefer hands-free documentation.
AI voice assistant for clinicians — dictate notes and control EHR by voice. Best for voice-driven clinical workflow.
Evidence-based clinical decision support — 12,000+ topics, drug info, AI search. The gold standard for clinical reference.