screen-reader-testing

Start with NVDA on Windows with Firefox or VoiceOver on macOS with Safari as minimum coverage. Navigate with arrow keys and Tab, verify page title, landmarks, headings, and form labels are announced, then test dynamic content and modals for proper announcements.

Minimum coverage is NVDA with Firefox, VoiceOver with Safari on macOS, and VoiceOver on iOS. Comprehensive coverage adds JAWS with Chrome, TalkBack with Chrome on Android, and Narrator with Edge.

Content injected into the DOM is not announced unless it uses a live region. Add role="status" with aria-live="polite" for status messages, or role="alert" with aria-live="assertive" for urgent errors that should interrupt current speech.

Use role="dialog" with aria-modal="true", aria-labelledby, and aria-describedby. Move focus into the modal on open, trap Tab and Shift+Tab within its focusable elements, close on Escape, and return focus to the triggering element.

Browse mode lets arrow keys read page content line by line, while focus mode passes keystrokes to interactive controls like form fields. NVDA switches automatically, and you can toggle manually with NVDA + Space.

No. Automated tools cannot verify what a screen reader actually announces, focus behavior, or live region timing. Test with real screen readers rather than simulators, and cover error states and dynamic content, not just the happy path.