Medical Transcription Typing Test
Real dictated report formats — SOAP notes, operative reports, discharge summaries.
Type the reports transcriptionists actually produce — histories, operative notes, discharge summaries and pathology reports, with the headings, doses and measurements intact. Built for transcriptionists, healthcare documentation specialists and speech-recognition editors.
What you will actually be typing
These are taken straight from the test. Notice what makes them different from ordinary medical prose: capitalised section headings, numbered plans, doses, laboratory values, and the particular rhythm of language that was spoken aloud before it was written down.
Vital signs are stable. Blood pressure is 124 over 78. Heart rate is 72 beats per minute. Respiratory rate is 16. Temperature is 98.6 degrees Fahrenheit. Oxygen saturation is 99 percent on room air. The patient appears comfortable and in no acute distress. He is alert and oriented to person, place, and time. He answers all questions appropriately and follows commands without difficulty. His skin is warm and dry, and his color is normal. The head is normal in shape with no signs of injury. The pupils are equal and react to light. The neck …
OPERATIVE NOTE: PREOPERATIVE DIAGNOSIS: Acute appendicitis. POSTOPERATIVE DIAGNOSIS: Same. PROCEDURE PERFORMED: Laparoscopic appendectomy. ANESTHESIA: General endotracheal. ESTIMATED BLOOD LOSS: Minimal. COMPLICATIONS: None. FINDINGS: An acutely inflamed appendix without evidence of perforation. The patient tolerated the procedure well and was transferred to the recovery room in stable condition with all sponge and instrument counts correct. DESCRIPTION OF PROCEDURE: After informed consent was obtained, the patient was brought to the operating room and placed in the supine position. General anesthesia was induced without difficulty, and the abdomen was prepped and draped in the usual sterile fashion. A …
PATHOLOGY REPORT: The specimen consists of a left modified radical mastectomy. Microscopic examination reveals an invasive ductal carcinoma measuring 2.3 centimeters in greatest dimension, histologic grade 2, with a Nottingham score of 6. The tumor is estrogen receptor positive at 95 percent, progesterone receptor positive at 80 percent, and human epidermal growth factor receptor 2 negative by immunohistochemistry and fluorescence in situ hybridization. Two of the fifteen axillary lymph nodes examined demonstrate metastatic carcinoma without extracapsular extension. GROSS DESCRIPTION: The specimen was received fresh, labeled with the patient's name and the designation left breast, and …
Who this test is for
This page is for people who produce clinical documents from dictation: working medical transcriptionists, healthcare documentation specialists, speech-recognition editors, and anyone preparing for a transcription service's pre-employment assessment. It is also useful if you are training toward the field and want to know whether the keyboard side of the work is within reach before you commit to a course.
One thing to be straight about at the outset: this is a typing test, not a transcription simulator. There is no audio. It trains the half of the job that happens at the keyboard — report structure, terminology, headings, doses and measurement detail. It cannot train your ear, your foot pedal, or your ability to work through a difficult accent. Those need real dictation practice.
| Durations | 1, 3, 5 or 10 minutes |
|---|---|
| Target speed | 50+ WPM net |
| Scoring formula | 5-character standard word; one standard word deducted per incorrect word |
| Backspace | Allowed (configurable) |
| Difficulty levels | Easy (clinical notes) · Medium (structured reports) · Hard (operative, pathology, consultation) |
| Passage corpus | 30 passages, 11,299 words, averaging 377 words per passage |
| Audio | None — this is a typing test, not a dictation exercise |
| Cost | Free, no login, results never leave your browser |
Dictated reports type differently from medical prose
People assume transcription is simply medical typing with harder words. It is not, and the difference is measurable. We analysed this corpus against the ordinary clinical prose used on our medical typing test. Terminology density is similar. What separates them is punctuation, capitalisation and digits.
| Level | Avg words each | Avg word length | Words 10+ chars | Digits per 100 chars |
|---|---|---|---|---|
| Easy | 374 | 4.84 chars | 6.7% | 0.24 |
| Medium | 379 | 5.48 chars | 12.2% | 0.8 |
| Hard | 376 | 6.05 chars | 16.7% | 0.57 |
That digit figure is the one worth dwelling on. Ordinary medical prose barely contains numbers; dictated reports are full of them — doses, vital signs, laboratory values, scores, dates, measurements. Numbers break touch-typing rhythm badly because they sit on a different row and carry no linguistic redundancy. You cannot infer "138" from context the way you can infer the ending of a familiar word. And a transposition that would be a harmless typo in prose becomes a dosing error in a report.
The capitalisation figure reflects the section headings that structure every dictated document. Reaching for shift several times per line, mid-flow, is its own distinct skill, and it is one that a general typing test never asks of you.
The report types you will transcribe
Dictated documents follow fixed structures. Knowing where the headings fall before the clinician reaches them is a large part of what makes an experienced transcriptionist fast — you are not reacting to the dictation so much as anticipating it.
| Report | When it is dictated | Standard sections | What makes it hard to type |
|---|---|---|---|
| History and physical (H&P) | On admission or before a procedure | Chief complaint, HPI, past medical and surgical history, medications, allergies, ROS, examination, assessment and plan | Long, heavily structured, dense medication and allergy detail |
| SOAP progress note | Each day of an admission, or each clinic visit | Subjective, Objective, Assessment, Plan | Short and repetitive in shape, but numeric — vitals, intake and output, labs |
| Operative report | Immediately after any procedure | Pre- and postoperative diagnosis, procedure, surgeon, anesthesia, findings, description, blood loss, counts | Instrument and anatomical vocabulary, strict heading order, dictated at speed |
| Discharge summary | At the end of an inpatient stay | Admission and discharge dates, diagnoses, hospital course, procedures, discharge medications, follow-up | The longest routine report; medication lists with doses are unforgiving |
| Consultation note | When one clinician formally advises another | Reason for consultation, history, examination, impression, numbered recommendations | Formal register, often opens and closes with courtesy language |
| Diagnostic report | Radiology, pathology, cardiology studies | Clinical indication, technique, findings, impression; pathology adds gross and microscopic description | Highest terminology density and measurement detail of any report type |
The passages in this test are drawn from all six categories, so practising here exposes you to each structure rather than to one repeated shape.
How transcription output is counted — and why it is not WPM
This test reports words per minute because that is what a typing test can measure. Your employer almost certainly will not. Transcription production is counted in lines, and understanding the unit matters because it determines your pay.
The industry standard is the visible black character: every printed letter, number, symbol and punctuation mark, excluding spaces, carriage returns, and all formatting such as bold, underline, italics and table structure. Divide the total visible character count by 65 and you have the line count. The older 65-character line was withdrawn as a standard in 1998, and the visible black character was recommended in its place specifically because it is the one counting method that every party to a contract can independently verify and reproduce [2].
Why the definition matters to your income
Counting methods that include spaces, or that count formatting characters, inflate the line count for the same amount of real work. Two services quoting an identical rate per line can pay materially differently if one counts visible black characters and the other counts gross characters including spaces. When comparing offers, the unit of measure is as important as the rate attached to it — ask which counting method is used before accepting a figure.
This has a practical consequence for practice. Speed on flowing prose is worth less to you than speed on the dense, punctuation-heavy, number-laden passages in the Hard set, because those produce more visible characters per minute of work.
The accuracy standard, and the nuance behind "98%"
Nearly every description of this field quotes a 98% accuracy requirement. The figure is real as an industry convention, but the detail behind it is more useful than the number, and almost nobody explains it.
AHDI quality guidance recommends that the target be expressed as a score of 98 rather than as 98 percent, for a specific reason: a quality assessment does not actually produce a percentage of accuracy [3]. It applies weighted deductions according to the severity of each error. A missing comma and an incorrect medication dose are not equivalent events, and any scoring model that treated them as equivalent would be measuring the wrong thing [5].
Errors are typically graded in tiers — critical errors that could affect patient care, major errors that alter meaning, and minor errors of style or punctuation. A single critical error can fail an otherwise clean report. This is why transcription employers weight accuracy more heavily than speed in pre-employment testing, and why the practice advice on this page keeps returning to numbers and drug names rather than to raw WPM.
Blanks and flags: what you never do
At some point the audio will be unclear. The clinician trails off, a word is masked by background noise, or an unfamiliar drug name arrives at speed through an accent you are still adjusting to. There is one correct response and one catastrophic one.
The correct response is to leave a blank and flag it for review under your employer's protocol, so a supervisor or the dictating clinician resolves it. Flagging is professional behaviour, not an admission of failure, and quality systems are designed on the assumption that it will happen.
The catastrophic response is to write down the closest plausible match. A guessed drug name, a guessed dose, or a guessed laterality enters a legal medical record looking exactly as confident as a verified one. Nothing downstream distinguishes it. This is the single most serious error available to someone in this role, and it is worth internalising before you ever touch live work.
Speech recognition changed the job — an honest look
You deserve a straight answer about the state of this field rather than an encouraging one. The U.S. Bureau of Labor Statistics projects employment of medical transcriptionists to decline 5 percent between 2024 and 2034, because speech recognition and natural language processing now let clinicians document encounters in real time, and because the same technology raises the output of the transcriptionists who remain [1].
The counterweight is that roughly 7,400 openings are still projected each year across the decade — essentially all of them arising from people leaving the occupation rather than from growth [1]. The work has not vanished. It has changed shape.
The change is from transcribing to editing. A recognition engine produces a draft and you correct it against the audio. That is a genuinely different skill, because machine errors are not human errors. A tired typist produces obvious mistakes. A recognition engine produces confident, grammatical, well-formed sentences that are wrong — a correct-sounding drug with the wrong name, the wrong laterality, a plausible dose that was never dictated. Fluent output passes a casual read, which makes editing a task of disciplined verification rather than of proofreading.
If you are entering the field now, the realistic framing is this: raw typing speed alone is a declining asset, while terminology knowledge, structural familiarity with report formats, and the judgement to catch a plausible-but-wrong machine output are appreciating ones. This test helps with the first two directly.
Credentials and equipment
AHDI offers two credentials for this field: the Registered Healthcare Documentation Specialist, aimed at those early in their career, and the Certified Healthcare Documentation Specialist for experienced practitioners working across multiple specialties [4]. Neither is legally required, but in a contracting market a credential is a reasonable way to distinguish an application.
On equipment, three items change daily output more than anything else. A foot pedal gives hands-free playback control so your fingers never leave the keyboard to rewind. Closed-back headphones materially reduce how often you need to replay unclear audio. And a text expander is the one most beginners underuse — mapping a few hundred long terms and standard phrases to short triggers removes a substantial share of the keystrokes from a working day, which matters directly when you are paid by the line.
The hardest terms in this test
Extracted from the Hard report set rather than chosen by hand. These are the longest distinct terms in the corpus and they account for a disproportionate share of the corrections most people make on this page.
- sacubitril-valsartan
- beta-hydroxybutyrate
- immunohistochemistry
- dextrose-containing
- receptor-neprilysin
- guideline-directed
- intermediate-grade
- multidisciplinary
- anti-inflammatory
- contrast-enhanced
- echocardiographic
- electrocardiogram
- impedance-derived
- insulin-dependent
- mineralocorticoid
- revascularization
- tomography-guided
- milliequivalents
- aminotransferase
- echocardiography
- gastroenterology
- hospitalizations
- over-the-counter
- hemodynamically
- hospitalization
- supplementation
- anticoagulation
- cerebrovascular
- discontinuation
- gastroenteritis
A four-week plan for transcription typing
Built around 20 to 30 minutes a day. Note that week two is about numbers rather than speed — that reflects where transcription errors actually cluster.
| Week | Focus | What to run | Move on when |
|---|---|---|---|
| Week 1 | Report shape | Easy passages, 5-minute runs — learn where headings fall | 95% accuracy; headings capitalised correctly without pausing |
| Week 2 | Numbers and doses | Medium passages, plus the alphanumeric drill | Zero transposed digits across a full 5-minute run |
| Week 3 | Terminology density | Hard passages — pathology, operative and consultation reports | Within 8 WPM of your Easy speed on Hard text |
| Week 4 | Sustained output | Hard passages, 10-minute runs, one attempt per session | 55+ WPM net at 98% accuracy held for the full 10 minutes |
For the numeric work in week two, the alphanumeric typing test and the numbers test drill exactly the patterns that appear in doses, laboratory values and record numbers. If your general speed is below about 40 WPM, work through how to type faster first — transcription text is a poor place to build fundamental technique.
How this differs from the medical typing test
The two pages are deliberately separate, and the corpus data above shows why the split is real rather than cosmetic.
This page covers the transcription profession: dictated report formats, line counting and the visible black character, turnaround and quality standards, working from audio, and the shift toward editing speech-recognition output.
The medical typing test covers typing clinical language as part of a healthcare job — terminology, spelling, abbreviations and electronic health record documentation for medical assistants, receptionists, scribes, nurses, coders and health information staff.
The simplest way to choose: if you would be typing what someone said, from a recording, you are in the right place. If you would be typing what you observed, entered or looked up, use the medical typing test instead.
Before a transcription assessment: a self-check
- You hold 55 WPM or better on Hard report passages, not just on Easy ones.
- Your accuracy stays above 98% across a full 10-minute run.
- You can type a medication line — drug, dose, route, frequency — without slowing or checking.
- Capitalised section headings no longer break your rhythm.
- Numbers are as fast as letters; no digit transpositions in a full run.
- You know your prospective employer's counting method and error-severity model.
- You have practised with actual dictation audio somewhere, since this test cannot supply it.
Frequently asked questions
What typing speed do you need for medical transcription?
Employers commonly advertise 60 to 80 WPM for transcription roles, which is markedly higher than the 35 to 45 WPM typical of general medical office work. The reason is structural rather than arbitrary: transcription pay is usually tied to output volume rather than hours, so your speed sets your earnings directly. There is no licensing body that publishes a required figure — the number comes from individual employers and transcription services, and it varies.
How is medical transcription work actually measured and paid?
Rarely in words per minute. Production is counted in lines, and the industry has moved toward the visible black character as the unit of measure — every printed letter, number, symbol and punctuation mark, with spaces, carriage returns and formatting excluded. Dividing the visible character count by 65 gives the line count. The older 65-character line standard was withdrawn in 1998; the visible black character was recommended as its replacement because it can be verified and reproduced by every party to a contract [2].
Is the 98% accuracy standard a real rule?
It is an industry convention rather than a regulation, and the detail behind it is more interesting than the number. AHDI quality guidance recommends expressing the target as a score of 98 rather than as 98 percent, because a quality assessment does not actually produce a percentage of accuracy — it applies weighted deductions for errors of differing severity, where a wrong drug dose is not equivalent to a missing comma [3][5]. Employers vary in the scoring model they apply.
Does this test include audio to transcribe from?
No, and that is worth being clear about. This is a typing test built on real dictated report text, not a transcription simulator. It trains the keyboard half of the job — report structure, terminology, headings, doses and measurement detail — and it will not train your listening, your foot pedal control, or your ability to work through a heavy accent or poor audio. For those you need actual dictation practice, ideally from a training program.
What do you do when you cannot make out a word in the dictation?
You never guess. The established practice is to leave a blank and flag it for review according to your employer's protocol, so that a supervisor or the dictating clinician resolves it. Inserting a plausible-sounding drug name or dose into a legal medical record because it approximately matches what you heard is the single most serious error a transcriptionist can make. Flagging is treated as correct professional behaviour, not as a failure.
Is medical transcription still a viable career?
Honestly, it is a contracting field. The Bureau of Labor Statistics projects employment of medical transcriptionists to decline 5 percent from 2024 to 2034, driven by speech recognition and natural language processing that let clinicians document in real time [1]. That said, roughly 7,400 openings are still projected each year, essentially all from workers leaving the occupation. The work that remains has shifted toward editing machine-generated drafts rather than typing from scratch.
What is the difference between transcribing and editing speech recognition output?
Transcribing means producing the document from audio yourself. Editing means the speech engine produces a draft and you correct it against the audio. Editing is now the more common arrangement and it is a genuinely different skill — the errors are not the ones a human makes. A recognition engine produces confident, well-formed, wrong words, so the risk is that a fluent-looking sentence passes review while containing the wrong drug or the wrong laterality.
Do I need a certification to work in medical transcription?
Not universally, but it helps in a shrinking market. AHDI offers two credentials: the Registered Healthcare Documentation Specialist for those early in their career, and the Certified Healthcare Documentation Specialist for experienced practitioners with multi-specialty exposure [4]. Many employers also run their own pre-employment test, which typically pairs a typing assessment with a transcription sample and a terminology component.
What equipment does the job actually require?
A foot pedal for hands-free playback control, closed-back headphones, and a text expander are the three that change your output most. The expander matters more than people expect — mapping a few hundred long terms and standard phrases to short triggers removes a large share of the keystrokes from a working day. Beyond that you need a reliable computer, a stable connection, and a private working space for confidentiality reasons.
Which report type should I learn first?
SOAP progress notes. They are short, they repeat the same four-part structure, and they appear in enormous volume, so the shape becomes automatic quickly. Move to history and physical reports next, since they introduce medication and allergy lists where accuracy pressure is highest. Leave operative and pathology reports until last — they carry the densest terminology and the least forgiving formatting.
How does confidentiality apply if I work from home?
Fully. Every report you handle is protected health information, and working remotely does not soften the obligation. In practice that means a workspace others cannot overlook, screens locked when you step away, no storage of report content on personal devices or cloud accounts, and no discussion of case detail with anyone outside the care relationship — including family. Most employers require a signed confidentiality agreement and periodic training.
Is this medical transcription typing test free?
Yes. No login, no signup, no limit on attempts. Scoring runs in your browser and your results are never uploaded to our servers.
Sources
Figures describing the passages on this page are computed from our own corpus and can be regenerated with scripts/corpus-stats.mjs. External claims are cited below. Typing speed ranges are drawn from commonly advertised roles and are labelled as such — no licensing or professional body publishes a required words-per-minute figure for this occupation.
- Occupational Outlook Handbook: Medical TranscriptionistsU.S. Bureau of Labor Statistics · link checked 21 Aug 2026
- The Visible Black Character: A Standard Measure for Transcribed ReportsAssociation for Healthcare Documentation Integrity (AHDI) · link checked 21 Aug 2026
- Healthcare Documentation Quality Assessment and Management Best PracticesAssociation for Healthcare Documentation Integrity (AHDI) · link checked 21 Aug 2026
- Credentialing: Registered and Certified Healthcare Documentation SpecialistAssociation for Healthcare Documentation Integrity (AHDI) · link checked 21 Aug 2026
- How Well Do You Measure Up? — on accuracy scoring in healthcare documentationFor The Record · link checked 21 Aug 2026