Creating accurate, synchronized captions for all video content.
🎬 Required for all UA video content
Why captions are required
Captions are required for all university video content under Title II ADA, Section 504, and UA policy. Beyond compliance, captions benefit:
- Deaf and hard of hearing viewers — Primary accessibility need
- Non-native English speakers — Supports language comprehension
- Students in noisy or quiet environments — Libraries, public transit
- Viewers with cognitive disabilities — Reinforces audio with text
- Search and findability — Captions are indexed and searchable
Research shows: 80% of people who use captions are not deaf or hard of hearing (3Play Media, 2019). Studies show captions improve comprehension and retention for all viewers.
Types of captions
| Type | Description | When to use |
|---|---|---|
| Closed captions (CC) | Can be turned on/off by viewer | Standard for most content |
| Open captions | Always visible, burned into video | Social media, signage displays |
| Subtitles | Translation for different languages | Multilingual content |
| SDH (Subtitles for Deaf/HoH) | Includes non-speech audio descriptions | When sound effects matter |
Caption quality standards
Captions must meet WCAG 2.1 AA requirements (Success Criterion 1.2.2). Quality captions are:
Accurate
- Verbatim speech (not paraphrased)
- Correct spelling of names and technical terms
- Proper grammar and punctuation
- Speaker identification when multiple speakers
- Target: 99% accuracy
Synchronized
- Captions appear within 1 second of speech
- Duration matches speech timing
- Line breaks at natural pause points
Complete
- All spoken content included
- Relevant non-speech sounds described: [music], [applause], [phone rings]
- Speaker changes indicated
- No missing or dropped words
Readable
- 2-3 lines maximum per caption frame
- 32-42 characters per line
- Sufficient display time to read (minimum 1 second per frame)
- Proper line breaks (don't split phrases awkwardly)
Captioning workflow at UA
Option 1: Panopto (recommended for courses)
UA's video platform with integrated captioning:
- Upload or record video in Panopto
- Enable ASR: Settings → Captions → Enable automatic captions
- Wait for processing (usually 1-2 hours for ASR)
- Edit captions: Click "Captions" in the video editor
- Review and correct all auto-generated text
- Publish when captions are accurate
Option 2: YouTube Studio
- Upload video to YouTube
- Go to Subtitles in YouTube Studio
- Wait for auto-generated captions
- Click Edit to review and correct
- Publish captions
Option 3: Professional captioning services
For high-volume or high-stakes content, use professional services:
| Service | Turnaround | Cost estimate | Best for |
|---|---|---|---|
| 3Play Media | 1-4 business days | $2-4/minute | High accuracy, large volumes |
| Rev | 12-24 hours | $1.50-2/minute | Fast turnaround |
| Verbit | Same day | $2-5/minute | Live events, lectures |
Contact accessibility@arizona.edu for department purchasing guidance.
Option 4: Manual captioning
For short videos or when other options aren't available:
- Amara: amara.org — Free web-based caption editor
- YouTube caption editor — Built into YouTube Studio
- Subtitle Edit: Free desktop software for Windows
Caption editing best practices
Common auto-caption errors to fix
- Names and proper nouns: "Dr. Smith" not "doctor smith"
- Technical terms: "WCAG" not "W cag" or "double-u cag"
- Homophones: "their/there/they're" — check context
- Numbers: Verify dates, statistics, phone numbers
- Acronyms: "UA" not "you a"
Formatting conventions
- Speaker identification: Use [Professor]: or >> Professor: for multiple speakers
- Sound effects: [music], [applause], [inaudible]
- Emphasis: Use italics for emphasis or foreign words
- Numbers: Spell out one through ten; use digits for 11+
Line breaking
Good:
The accessibility guidelines require captions on all videos.
Poor:
The accessibility guidelines require captions on all videos.
Break at natural phrase boundaries, not in the middle of thoughts.
Live event captioning
For live events, webinars, and synchronous classes:
Zoom automatic captions
- Enable in meeting settings
- Quality varies (70-85% accuracy)
- Suitable for informal meetings
- Not sufficient for formal accommodations
Professional CART (Communication Access Realtime Translation)
- Human captioner provides 98%+ accuracy
- Required for formal accommodations
- Schedule through DRC for student accommodations
- Book 2+ weeks in advance for events
Contact uadrc@arizona.edu to arrange CART services.
Transcripts
In addition to captions, provide downloadable transcripts for:
- Audio-only content (podcasts)
- Long-form videos
- Content where users may want to search or reference
Transcript best practices
- Include speaker names
- Add timestamps for long content
- Use headings for sections/topics
- Note significant non-speech audio
- Provide in accessible format (HTML or accessible Word/PDF)
Audio description
Audio description narrates important visual information for blind and low-vision viewers. Required when:
- Visual content conveys information not in the audio track
- Demonstrations, charts, or on-screen text
- Videos where action is important to understanding
When audio description is needed
| Content type | Audio description needed? |
|---|---|
| Talking head lecture | Usually no (if slides are described verbally) |
| Software demo | Yes (describe on-screen actions) |
| Lab demonstration | Yes (describe physical actions) |
| Chart or graph shown | Yes (describe data and trends) |
| Interview | Usually no (unless visual context matters) |
Providing audio description
- Integrate into script: Narrate what you're showing as you present
- Extended description: Pause video to describe complex visuals
- Separate track: Create alternative version with AD
- Text description: Provide written description of visual content
Caption quality checklist
- All speech is captioned verbatim
- Speaker identification included
- Names and technical terms spelled correctly
- Non-speech sounds described [music], [applause]
- Captions synchronized with audio
- Line breaks at natural phrases
- 99%+ accuracy achieved
- Captions display long enough to read
Resources
- DCMP Captioning Key — Industry standard guide
- 3Play Media Resources — Best practices and research
- Panopto Workflow Guide
- Teaching & Learning Hub