Skip to main content
Understandable WCAG 3.1.6

3.1.6 Pronunciation

A mechanism is available for identifying specific pronunciation of words where meaning is ambiguous without knowing the pronunciation.

Level AAA Minor WCAG 2.0 (new) WCAG 2.1 WCAG 2.2

What this rule means

WCAG 3.1.6 requires that when the meaning of a word depends on its pronunciation (homographs), a mechanism is provided to clarify. For example, "lead" can mean a metal (led) or to guide (leed). Without pronunciation guidance, meaning is lost.

This applies to languages where written form does not fully determine pronunciation. In English, homographs are common. In languages with phonetic writing systems, this is less of an issue.

Why it matters

Screen readers may pronounce homographs incorrectly, changing the meaning entirely. Users with reading disabilities may also struggle to determine the correct pronunciation from context alone.

Providing pronunciation guides improves comprehension and ensures screen readers convey the correct meaning.

Related axe-core rules

There are no automated axe-core rules for this criterion. It requires manual review of content for ambiguous pronunciation.

How to test

  • Identify words with multiple pronunciations that change meaning (homographs).
  • Check whether context alone resolves the ambiguity.
  • Verify pronunciation guidance is provided where context is insufficient.

How to fix

<!-- Using ruby annotation for pronunciation -->
<p>The <ruby>lead<rp>(</rp><rt>led</rt><rp>)</rp></ruby>
   pipes were replaced last year.</p>

<!-- Using parenthetical pronunciation -->
<p>The lead (rhymes with "bed") content in old paint
   is a health hazard.</p>

<!-- Providing SSML hints for screen readers -->
<!-- Note: limited support, mainly in specialized applications -->
<span data-ssml='<phoneme alphabet="ipa" ph="lɛd">lead</phoneme>'>
  lead
</span>

Common mistakes

  • Ignoring homographs entirely and relying on context that may be ambiguous.
  • Using pronunciation guides inconsistently throughout the content.
  • Providing pronunciation only in audio form, excluding users who are deaf or hard of hearing.

Resources