Only 723 people hold NCRA’s Certified Realtime Captioner credential, and just 3% of the association’s reporter members work as CART captioners, even though 20.5% of US undergraduates report a disability and state courts recorded 70 million filings in 2024 (NCRA Statistics 2026; NCES Digest of Education Statistics; NCSC Court Statistics Project). The broader occupation that supplies realtime writers, court reporters and simultaneous captioners, holds 19,900 jobs with 0% projected growth through 2035 (BLS Occupational Outlook Handbook). Meanwhile, 90.9% of college disability offices in a recent National Deaf Center survey said deaf students requested CART, and 22.6% admitted speech-to-text requests sometimes or often went unfilled. We aggregated data from the Bureau of Labor Statistics, the National Court Reporters Association, the National Center for Education Statistics, the National Deaf Center, the National Center for State Courts, the US Department of Justice, WHO, NIDCD and peer-reviewed research in PNAS, Language and Frontiers in Digital Health. For the machine side of courtroom transcription, see our companion roundup on courtroom speech recognition accuracy.
TL;DR
- 19,900 court reporter and simultaneous captioner jobs in the US, 0% projected growth for 2025-35 (BLS Occupational Outlook Handbook 2025)
- $72,420 median annual wage, May 2025 (BLS Occupational Outlook Handbook 2025)
- 723 Certified Realtime Captioners and 2,011 Certified Realtime Reporters (NCRA Statistics 2026)
- 11% pass rate on 304 CRR realtime skills tests in 2025 (NCRA Statistics 2026)
- Average NCRA reporter member is 56 years old, 88% of members are female (NCRA Statistics 2026)
- 70 million state court filings in 2024, up 4% from 2023 (NCSC Court Statistics Project)
- $3.70 per page for a single realtime feed in federal courts (US Courts, Federal Court Reporting Program)
- 20.5% of US undergraduates, 3,478,000 students, reported a disability in 2019-20 (NCES Digest Table 311.10)
- 90.9% of surveyed colleges said deaf students requested CART (National Deaf Center DSS Report, 2023-24 survey)
- 22.6% of colleges said speech-to-text requests sometimes or often went unfilled (National Deaf Center DSS Report, 2023-24 survey)
- 0.35 vs 0.19 average ASR word error rate for Black vs white speakers (ASR racial disparities study, PNAS 2020)
- About 1% of Whisper transcriptions contained hallucinated text, 38% of those harmful (Careless Whisper study, FAccT 2024)
1. The Realtime Workforce: Jobs and Pay
CART providers are drawn from the same stenographic labor pool as court reporters, so the BLS occupation “court reporters and simultaneous captioners” is the closest official headcount for the people who can write realtime. That pool is small and flat: 19,900 jobs nationally with a 0% projected change from 2025 to 2035, according to the BLS Occupational Outlook Handbook.
The zero-growth projection hides real churn. BLS still expects about 1,800 openings per year, almost all from retirements and exits rather than expansion. The handbook also flags a split future: AI tools are dampening demand for simultaneous captioners, while an aging population may increase the need for CART. Self-employed workers make up 33% of the occupation, local government 24%, state government 20% and business support services 19%, which means a large share of CART capacity sits with freelancers and agencies rather than inside courts or colleges.
| Metric | Value | Source |
|---|---|---|
| Court reporter and simultaneous captioner jobs | 19,900 (2025) | BLS Occupational Outlook Handbook 2025 |
| Median annual wage | $72,420 ($34.82 per hour), May 2025 | BLS Occupational Outlook Handbook 2025 |
| Lowest 10% earned less than | $42,130 | BLS Occupational Outlook Handbook 2025 |
| Highest 10% earned more than | $130,560 | BLS Occupational Outlook Handbook 2025 |
| Median wage, local government | $78,430 | BLS Occupational Outlook Handbook 2025 |
| Median wage, state government | $78,300 | BLS Occupational Outlook Handbook 2025 |
| Median wage, business support services | $58,760 | BLS Occupational Outlook Handbook 2025 |
| Projected employment change, 2025-35 | 0% (about 1,800 openings per year) | BLS Employment Projections via OOH |
Context note: the government premium is notable. Reporters employed by local and state government earn $19,670 more at the median than those in business support services ($78,430 minus $58,760, our calculation), the segment where many freelance CART agencies sit.
2. Credentials and Testing: Why Realtime Is Scarce
A certified CART writer must produce near-perfect text at conversational speed with no chance to edit. NCRA’s Certified Realtime Reporter exam demands a five-minute testimony segment at 200 words per minute with 96% accuracy, and the Certified Realtime Captioner exam requires 180 words per minute at 96% accuracy on literary matter, plus a 50-question written test with a 70% passing score.
The pass rates explain the scarcity. Only 11% of 304 CRR skills tests were passed in 2025, against 29% for the RPR-L skills test (421 tests) and 61% for the RMR-L (41 tests), per NCRA Statistics. Across all credentials, NCRA awarded 205 new certifications in 2025. Since October 1, 2025, the education component is no longer required for CRC candidates, a change that removes one hurdle from the entry-level captioning credential.
| Metric | Value | Source |
|---|---|---|
| Registered Professional Reporter (RPR) holders | 4,776 | NCRA Statistics 2026 |
| Certified Realtime Reporter (CRR) holders | 2,011 | NCRA Statistics 2026 |
| Certified Realtime Captioner (CRC) holders | 723 | NCRA Statistics 2026 |
| Registered Merit Reporter (RMR) holders | 1,317 | NCRA Statistics 2026 |
| Registered Diplomate Reporter (RDR) holders | 497 | NCRA Statistics 2026 |
| CRR exam standard | 200 wpm, 96% accuracy, 5 minutes, no editing | NCRA CRR certification page 2026 |
| CRC exam standard | 180 wpm, 96% accuracy, 5 minutes | NCRA CRC certification page 2026 |
| CRR skills test pass rate, 2025 | 11% (304 tests) | NCRA Statistics 2026 |
Outlier note: the written knowledge tests are far easier to clear than the realtime skills tests. The RPR written test had a 63% pass rate on 255 attempts in 2025, more than five times the CRR skills pass rate.
3. Membership, Aging and the Shortage Pipeline
The supply problem is demographic. NCRA’s March 2026 snapshot puts the average reporter member age at 56, with 88% of members female and only 8% of members students. Of 7,630 reporter members, 46% are freelancers and 28% are officials, while CART captioners account for 3%, broadcast captioners 1% and combined CART and broadcast another 1%. Applying the 3% share to 7,630 reporter members gives roughly 229 dedicated CART captioners inside NCRA (our calculation), a thin bench for a national accommodation obligation.
The training pipeline is narrow too: NCRA lists 19 approved court reporting programs. The last full industry forecast is now dated. Most recent available data: the 2013-14 Court Reporting Industry Outlook by Ducker Worldwide for NCRA, which projected demand of 33,200 against supply of 27,700 by 2018, a gap of 5,500 positions. It counted about 32,000 stenographic court reporters, fewer than 1,000 working exclusively in captioning, and some states with application declines “as high as an 85 percent decrease” over five years.
| Metric | Value | Source |
|---|---|---|
| NCRA total members | 10,550 (March 18, 2026) | NCRA Statistics 2026 |
| NCRA reporter members | 7,630 | NCRA Statistics 2026 |
| CART captioners among reporter members | 3% | NCRA Statistics 2026 |
| Average reporter member age | 56 | NCRA Statistics 2026 |
| NCRA-approved court reporting programs | 19 | NCRA Statistics 2026 |
| Projected shortfall by 2018 | 5,500 positions (33,200 demand vs 27,700 supply) | NCRA / Ducker Industry Outlook 2013-14 |
| Reporters expected to retire within 20 years | About 70% | NCRA / Ducker Industry Outlook 2013-14 |
| Largest projected state gap, 2018 | California, 2,320 positions | NCRA / Ducker Industry Outlook 2013-14 |
Context note: in 2013 the Ducker report put the median court reporter age at 51 versus 42 for all occupations. NCRA’s current figure is an average for members, not a median for the whole occupation, so the two numbers are not a strict trend line, but both point to an aging workforce.
4. Courts and the Legal Mandate
Demand for realtime in the justice system starts with volume. The NCSC Court Statistics Project recorded 70 million state court filings in 2024, a 4% increase from 2023 but 27% below 2012. Every hearing where a deaf or hard-of-hearing party, juror or witness needs access becomes a potential CART assignment, and under ADA Title II courts must give “primary consideration” to the aid the person requests, according to DOJ effective communication guidance. DOJ describes CART as “a service similar to court reporting” that is “particularly useful for people who are deaf or have hearing loss but do not use sign language.”
Federal courts price realtime explicitly. The US Courts Federal Court Reporting Program sets a maximum of $3.70 per page for one realtime feed, falling to $2.55 for two to four feeds and $1.80 for five or more. Digital access rules are also moving: DOJ’s April 20, 2026 interim final rule extended Title II web accessibility deadlines to April 26, 2027 for entities serving 50,000 or more people and April 26, 2028 for smaller ones, with WCAG 2.1 Level AA as the standard (Federal Register 2026-07663).
| Metric | Value | Source |
|---|---|---|
| State court filings, 2024 | 70 million | NCSC Court Statistics Project 2024 |
| Change in filings vs 2023 | +4% | NCSC Court Statistics Project 2024 |
| Change in filings vs 2012 | -27% | NCSC Court Statistics Project 2024 |
| Federal realtime feed, one feed | $3.70 per page | US Courts Federal Court Reporting Program |
| Federal realtime feed, five or more feeds | $1.80 per page | US Courts Federal Court Reporting Program |
| Federal 2-hour (hourly) transcript, original | $8.70 per page | US Courts Federal Court Reporting Program |
| Federal 30-day transcript, original | $4.40 per page | US Courts Federal Court Reporting Program |
| Title II web compliance, entities of 50,000+ | April 26, 2027 (extended) | DOJ Interim Final Rule, Federal Register 2026 |
Context note: the ADA rule on who chooses the aid differs by title. Title II entities (state and local government, including state courts) must give primary consideration to the person’s choice, while Title III businesses are only “encouraged to consult” with the person (DOJ, ADA Requirements: Effective Communication, updated 2020).
5. Students Who Need Access: Classroom Demand
The classroom side of CART demand is larger than the deaf population alone. 20.5% of US undergraduates, about 3,478,000 of 16,936,000 students, reported a disability in 2019-20, and 10.7% of postbaccalaureate students did, per NCES Digest Table 311.10. NCES’s definition includes “deafness or serious difficulty hearing” alongside vision, cognitive and mobility difficulties. Most recent available data: NPSAS 2019-20.
In K-12, 7.5 million students ages 3 to 21, or 15% of public school students, received IDEA services in 2022-23, and 67% of school-age IDEA students spent 80% or more of the day in general classes, up from 61% in 2012 (NCES Condition of Education). Students served for hearing impairments fell from 77,000 in 2012-13 to 70,000 in 2022-23 (NCES Digest Table 204.30), but inclusion means more of them sit in mainstream classrooms where spoken lectures need text access. At the college level, the National Deaf Center reports that 1.4% of college students are deaf, and only 22% of deaf people hold a bachelor’s degree or higher, 15.7% lower than hearing peers.
| Metric | Value | Source |
|---|---|---|
| Undergraduates reporting a disability, 2019-20 | 20.5% (3,478,000) | NCES Digest Table 311.10 |
| Postbaccalaureate students reporting a disability | 10.7% | NCES Fast Facts, NPSAS 2019-20 |
| Students served under IDEA, 2022-23 | 7.5 million (15% of public school students) | NCES Condition of Education (2022-23 data) |
| IDEA students with hearing impairments, 2022-23 | 70,000 (vs 77,000 in 2012-13) | NCES Digest Table 204.30 |
| College students who are deaf | 1.4% | National Deaf Center, citing NCES 2023 |
| Deaf people enrolled in postsecondary education | 5% (vs 9.9% of hearing people) | National Deaf Center, citing ACS 2024 |
| Deaf college students with an additional disability | 45.8% | National Deaf Center, citing NCES 2023 |
| Deaf students receiving more than one accommodation | 60% | National Deaf Center, citing DPAIS 2023 |
Context note: the hearing population that will need text access keeps growing with age. NIDCD estimates disabling hearing loss affects 22% of adults 65 to 74 and 55% of those 75 and older, and CDC puts hearing disability at 6.2% of US adults (CDC Disability Impacts All of Us). For the demographic detail, see our roundup on age-related hearing loss.
6. How Colleges Deliver CART Today
Requests are nearly universal; supply is not. In the National Deaf Center survey of disability services professionals, 65 offices (one per institution) responded between April 2023 and May 2024. 90.9% reported that deaf students requested CART services, 73.9% provided speech-to-text services in the last year, and 22.6% acknowledged that speech-to-text requests “sometimes or often went unfilled.”
The comparison with sign language interpreting is revealing. 89.2% of institutions provide interpreting and 72.8% said they were rarely or never unable to fill an interpreting request, yet only 24.8% employ staff interpreters. Speech-to-text has gone heavily remote: on average only 41% of those services are provided in person. Policy consistency lags as well, with 53.5% of respondents saying it was unlikely that captioning policies were being consistently applied across campus.
| Metric | Value | Source |
|---|---|---|
| Institutions where deaf students requested CART | 90.9% | National Deaf Center DSS Report, 2023-24 survey |
| Institutions providing speech-to-text in the last year | 73.9% | National Deaf Center DSS Report, 2023-24 survey |
| Speech-to-text requests sometimes or often unfilled | 22.6% of institutions | National Deaf Center DSS Report, 2023-24 survey |
| Speech-to-text services delivered in person | 41% on average | National Deaf Center DSS Report, 2023-24 survey |
| Institutions providing captioned media | 81.5% | National Deaf Center DSS Report, 2023-24 survey |
| Captioning policies unlikely to be applied consistently | 53.5% of respondents | National Deaf Center DSS Report, 2023-24 survey |
| Institutions providing sign language interpreting | 89.2% | National Deaf Center DSS Report, 2023-24 survey |
| Institutions with staff interpreters | 24.8% | National Deaf Center DSS Report, 2023-24 survey |
Caveat: 65 responses is a small, self-selected sample, so these shares describe engaged disability offices rather than all US institutions. The direction is still clear: CART is the accommodation most often requested and least reliably staffed. Video captioning patterns outside the classroom are covered in our short-form video caption usage roundup.
7. Human CART vs Automatic Captions
Colleges are already substituting software: in the same National Deaf Center survey, 25.6% of institutions cited auto-captions and another 25.6% auto-captions with manual corrections among their captioning methods, and 65.9% use AI technologies for note-taking services. The research record shows why that swap is risky for legal and classroom access. A 2020 PNAS study of five commercial ASR systems (Amazon, Apple, Google, IBM, Microsoft) across 19.8 hours of interviews found an average word error rate of 0.35 for Black speakers versus 0.19 for white speakers.
Hallucination is a separate failure mode that human writers do not have. The Careless Whisper study (FAccT 2024) found that about 1% of audio transcriptions contained entirely fabricated phrases, and 38% of those hallucinations included explicit harms. Humans are not error-free either: court reporters are certified at either 95% or 98% accuracy depending on the certifying body, yet Philadelphia reporters failed to meet that level on African American English in an experimental study (Jones et al., Language 2019).
| Metric | Value | Source |
|---|---|---|
| Average ASR word error rate, Black vs white speakers | 0.35 vs 0.19 | ASR racial disparities study, PNAS 2020 |
| Whisper transcriptions with fabricated text | About 1% | Careless Whisper study, FAccT 2024 |
| Hallucinations containing explicit harms | 38% | Careless Whisper study, FAccT 2024 |
| Word error rate of ASR apps, English dialogue | 19-34% | ASR apps evaluation, Frontiers in Digital Health 2022 |
| Word error rate of ASR apps, Dutch dialogue | 25-66% | ASR apps evaluation, Frontiers in Digital Health 2022 |
| Court reporter certification accuracy standards | 95% or 98% | Jones et al., Language 2019 |
| Colleges using auto-captions (unedited) | 25.6% | National Deaf Center DSS Report, 2023-24 survey |
| Colleges using AI for note-taking services | 65.9% | National Deaf Center DSS Report, 2023-24 survey |
Context note: accuracy is only one dimension. A September 2026 preprint by researchers including Raja Kushalnagar and Christian Vogler, with 216 deaf and hard-of-hearing participants rating 70 TV clips, found that viewers struggled with the typical 7 to 12 second delays of live TV captions and that Word Error Rate correlated “far less” with satisfaction for ASR captions (arXiv 2609.11408). The Frontiers study concluded that in quiet conditions ASR apps “performed similarly to listeners with a moderate hearing loss.” Market-level context sits in our captioning and subtitling market roundup.
Summary: CART Legal and Classroom Statistics by the Numbers
| Metric | Value | Source |
|---|---|---|
| Court reporter and simultaneous captioner jobs | 19,900 (2025) | BLS Occupational Outlook Handbook 2025 |
| Median annual wage | $72,420 (May 2025) | BLS Occupational Outlook Handbook 2025 |
| Projected employment change, 2025-35 | 0%, about 1,800 openings per year | BLS Occupational Outlook Handbook 2025 |
| Certified Realtime Captioner holders | 723 | NCRA Statistics 2026 |
| Certified Realtime Reporter holders | 2,011 | NCRA Statistics 2026 |
| CRR skills test pass rate, 2025 | 11% of 304 tests | NCRA Statistics 2026 |
| CART captioners among NCRA reporter members | 3% | NCRA Statistics 2026 |
| Average NCRA reporter member age | 56 | NCRA Statistics 2026 |
| Projected shortfall by 2018 (most recent forecast) | 5,500 positions | NCRA / Ducker Industry Outlook 2013-14 |
| State court filings, 2024 | 70 million | NCSC Court Statistics Project 2024 |
| Federal realtime feed, one feed | $3.70 per page | US Courts Federal Court Reporting Program |
| Title II web compliance, large entities | April 26, 2027 | DOJ Interim Final Rule 2026 |
| Undergraduates reporting a disability | 20.5% (2019-20) | NCES Digest Table 311.10 |
| IDEA students with hearing impairments | 70,000 (2022-23) | NCES Digest Table 204.30 |
| Deaf people with a bachelor’s degree or higher | 22% | National Deaf Center, citing ACS 2024 |
| Colleges where deaf students requested CART | 90.9% | National Deaf Center DSS Report, 2023-24 survey |
| Colleges with unfilled speech-to-text requests | 22.6% | National Deaf Center DSS Report, 2023-24 survey |
| ASR word error rate, Black vs white speakers | 0.35 vs 0.19 | ASR racial disparities study, PNAS 2020 |
| Whisper transcriptions with hallucinations | About 1% | Careless Whisper study, FAccT 2024 |
| People worldwide needing rehabilitation for disabling hearing loss | Over 430 million | WHO Deafness and Hearing Loss Fact Sheet 2026 |
Methodology and Sources
- BLS Occupational Outlook Handbook: Court Reporters and Simultaneous Captioners: jobs, wages (May 2025), industry shares and 2025-35 projections.
- NCRA Statistics: membership (March 18, 2026), credential counts, 2025 test pass rates, approved programs.
- NCRA certification pages for the CRR and CRC: speed and accuracy standards.
- NCRA / Ducker Worldwide 2013-14 Court Reporting Industry Outlook: shortage projections, retirement and enrollment data.
- NCSC Court Statistics Project: 2024 state court filings.
- US Courts Federal Court Reporting Program: maximum transcript and realtime rates.
- DOJ ADA Requirements: Effective Communication and DOJ Title II web rule fact sheet, plus the April 2026 interim final rule.
- NCES Digest Table 311.10, NCES Fast Facts on students with disabilities, NCES Digest Table 204.30 and NCES Condition of Education.
- National Deaf Center Deaf College Student Data and DSS Report.
- WHO Deafness and Hearing Loss Fact Sheet (March 2026), NIDCD Quick Statistics About Hearing and CDC Disability Impacts All of Us.
- Peer-reviewed and preprint research: ASR racial disparities study, PNAS 2020, Careless Whisper, FAccT 2024, Jones et al., Language 2019, ASR apps evaluation, Frontiers in Digital Health 2022 and Thompson et al., arXiv 2026.
- Data watch: NCRA’s About page describes “more than 12,000 members” and its Captioning Matters page “approximately 12,000,” while its statistics page lists 10,550 members as of March 18, 2026; we use the dated statistics figure. The only national shortage forecast (Ducker 2013-14) is more than ten years old and its 2018 horizon has passed, so treat it as historical context. NCES disability shares come from NPSAS 2019-20, and the next NPSAS release should update them. The NCRA 3% CART share covers association members only, not all US providers. The BLS OEWS detail page for the occupation could not be read during research, so wage figures come from the Occupational Outlook Handbook. The National Deaf Center survey has 65 respondents and the September 2026 caption-metrics paper is a preprint not yet peer reviewed. WHO also reports that 95.1 million children aged 5 to 19 live with hearing loss and projects nearly 2.5 billion people with some degree of hearing loss by 2050, which signals rising long-term demand for text access.
Last updated: October 4, 2026. We update this roundup quarterly, and the next refresh is expected when BLS publishes its May 2026 OEWS estimates and NCRA posts its 2026 certification testing results.