Understanding Deaf and Hard-of-Hearing Users
Approximately 48 million people in the United States have hearing loss, with millions more worldwide. Audio accessibility—providing alternatives to audio content—is essential for deaf and hard-of-hearing users. Captions, transcripts, sign language interpretation, and visual alternatives ensure audio content is accessible to all users. The deaf and hard-of-hearing community encompasses diverse experiences. Some users are completely deaf from birth or early childhood; others lost hearing later in life. Some use spoken language; others use sign language as primary communication. Hearing aid users might have better hearing in quiet environments but struggle in noisy settings. For website design, the key principle is this: audio content must have accessible alternatives. If your site includes videos, podcasts, webinars, or other audio content, that content must be accessible to users who cannot hear it. Unlike visual accessibility (which affects 1 in 4 adults at some point), hearing loss is often invisible. Users might not disclose hearing loss, making it easy for organizations to overlook audio accessibility needs.
Core Audio Accessibility Elements
Closed captions (CC) transcribe dialogue and sound effects from video content. They include not just spoken words but also [ambient traffic], [door slams], [music playing]—giving deaf viewers complete audio understanding. Good captions include: Audio descriptions provide narrated descriptions of visual content. For videos, audio descriptions are spoken during natural pauses in dialogue, describing on-screen actions users cannot hear. Example: During a instructional video, audio description might say: "The technician carefully removes the circuit board from the device" during the scene showing this action. Audio descriptions allow blind users to understand visual content; they also benefit low-vision users and users in environments where they cannot watch closely. Transcripts are written text versions of audio content. For podcasts, webinars, or recorded videos, provide full transcripts. Transcripts provide: For important video content, consider sign language interpretation. Many deaf users prefer sign language to captions. Sign language interpreters appear in a window during videos, interpreting content. While more resource-intensive than captions, sign language interpretation benefits deaf users who are Deaf (capital D—culturally deaf, using sign language as primary communication). Some content relies on audio cues users must hear: error notifications, confirmation sounds, or alerts. For such content, provide visual equivalents: error messages appear on screen, confirmation icons flash, or alerts are visually highlighted.
All dialogue and spoken content
Sound effects in brackets: [phone ringing], [laughter]
Music descriptions: [upbeat jazz music]
Speaker identification: [JOHN]: "Let's begin"
Accurate timing synchronized with video
Text access to audio-only content (podcasts)
Searchable content (users can find specific topics)
Better SEO (search engines index text, not audio)
Multiple accessibility benefit (screen readers can read transcripts)
Designing Video for Accessibility
Video players themselves must be accessible: Poor captions are sometimes worse than no captions. Guidelines for caption quality: Captions appear at video bottom by default, but position should be adjustable. Some caption formats (open captions burned into video) cannot be repositioned; closed captions (separate, toggleable) should be adjustable in position. Transcripts should:
Caption toggle: Clear, obvious button to turn captions on/off
Caption visibility: Captions must be readable with 7:1 contrast, large enough to read, and not blocking important video content
Transcript link: Direct link to transcript near player
Keyboard controls: Play/pause, volume, closed captions accessible via keyboard
Focus management: Keyboard users can access all controls
Accuracy: Captions must accurately represent spoken content
Completeness: All dialogue and relevant sounds included
Timing: Captions appear and disappear synchronized with audio
Speaker identification: Clear who is speaking, especially in multi-speaker content
Sound descriptions: Important sound effects and music included
Readability: Text large enough, readable fonts, sufficient contrast
Be clearly labeled and easily accessible near video player
Include speaker identification for clarity in multi-speaker content
Include timestamps for reference
Describe important sound effects and music
Include text formatting for readability (paragraphs, speakers on new lines)
Designing for Audio-Only Content (Podcasts)
Podcasts and audio-only content must have transcripts. Provide complete transcripts of all podcast episodes, clearly linked from episode pages. Podcast transcripts should: Place transcripts on the same page as audio player. Users should not need to search for transcripts; they should be obvious. Even sponsored segments and advertisements in podcasts should be transcribed. Exclude only truly incidental background audio.
Include all spoken content accurately
Identify speakers clearly
Include timestamps every few minutes
Include descriptions of significant sound effects or music
Be searchable (PDF transcripts are less searchable than HTML)
Live Events and Webinars
For live streams and webinars, real-time CART (Communication Access Realtime Translation) captioning provides live captions. Professional CART services employ trained captioners who listen and type captions in real-time, typically with 99% accuracy. When hosting webinars or live events: After webinars conclude, captions should be added to recordings before publishing. This requires either recording the CART caption feed or adding captions in post-production.
Budget for professional CART captioning
Provide platform for captions display
Allow speakers to review captions (some services provide speaker console)
Record webinars for later captioning and archiving
Audio Descriptions in Detail
Audio descriptions are most important for: Audio descriptions are less critical for videos that are primarily talking-head (someone speaking to camera) where dialogue covers all necessary information. Good audio descriptions: Professional audio description services employ trained describers who watch video and write descriptions, then narrators record those descriptions, which are edited into video during natural pauses. For organizations with limited budgets, AI-assisted tools are emerging that automatically identify key visual moments and suggest descriptions, which professionals then refine.
Instructional videos showing procedures or techniques
Training videos with visual demonstrations
Narrative videos where visual storytelling is central
Product demonstrations
Are concise but complete
Fit naturally into content pauses
Use clear, simple language
Describe important visual details, not obvious elements
Are objective, not subjective or interpretive
Testing Audio Accessibility
Watch videos with captions on, audio off. Verify: Read transcripts independently. Verify: If possible, involve deaf users in testing. Deaf users identify caption quality issues that hearing testers miss and provide invaluable feedback on accessibility effectiveness.
All dialogue is captured accurately
Sound effects are described
Captions are visible and readable
Timing is synchronized properly
Speaker identification is clear
Transcripts match spoken content
Sound descriptions are included where relevant
Transcripts are easy to read and navigate
Timestamps help navigation
Common Mistakes to Avoid
Auto-generated captions only: Automated captions are often inaccurate and miss sound effects. They're better than no captions but should be reviewed and corrected.
No transcript: Videos may have captions but no transcript for easy reference or searching.
Burned-in captions: Open captions burned into video cannot be turned off or repositioned. Use closed captions.
Missing speaker identification: Multi-speaker content without clear speaker IDs is confusing.
No sound effect descriptions: Captions showing only dialogue miss important audio context.
Unreadable caption appearance: Small captions, low contrast, or covering important video content reduce accessibility.
No audio description consideration: Assuming captions are sufficient when video requires visual descriptions.
Content Creator Responsibilities
Organizations should establish accessibility requirements for content creators:
All videos must have captions before publishing
Captions must be reviewed for accuracy, not just auto-generated
Transcripts must be provided for audio-only content
Instructional videos should include audio descriptions
Caption and transcript files should be organized and linked properly