Automated captions are not compliant alone. Manual review and correction are mandatory for all Florida State University multimedia content.
Compliance Standards by Content Type
| Content Type | WCAG Requirement | How to Achieve Compliance |
|---|---|---|
| Pre-Recorded Video | Captions & Audio Descriptions | Mandatory: Create a full script, review auto-generated captions for 100% accuracy, and upload the final VTT/SRT file. |
| Live Events (Zoom/Teams) | Live Captions | Use the platform’s live captioning feature; ensure a mechanism is in place for accommodations if quality is insufficient. |
| Audio-Only (Podcasts) | Transcripts | Provide a separate, clean, scrollable text page containing the full content of the audio. |
| YouTube Uploads | Text Alternative | Action: Paste the full transcript into the YouTube description box with the heading: "Full Text Transcript." |
Multimedia Best Practices
Captioning Essentials
- Accuracy: Aim for 99% accuracy. Correct all misspellings, proper nouns, and punctuation.
- Synchronization: Captions must appear exactly when the audio starts.
- Speaker ID: Identify multiple speakers (e.g., [PROFESSOR SMITH]:).
- Sound Effects: Include non-speech audio in brackets (e.g., [applause], [upbeat music]).
- Placement: Leave the bottom 20% of the screen clear for readability.
Advanced Workflows (Canva & AI)
- Auto-Captioning: Use the Canva Auto-Caption tool or Canva Video Editor to start, then manually edit.
- Inclusive Design: Use AI to resize and translate content for broader access.
- Training: Review the Canva Accessibility Tutorial for visual walkthroughs.
AI Tools for Compliance
Copy and paste these prompts into Gemini, ChatGPT, or Co-Pilot along with your raw text to transform messy auto-transcripts into accessible formats.
Prompt 1: The "Caption Cleaner"
Act as a WCAG 2.1 Level AA Accessibility Specialist. I am going to paste a raw transcript of a video below. Please rewrite it into a caption-ready format following these strict rules: 1. Speaker Identification: Identify speakers in UPPERCASE (e.g., PROFESSOR:). 2. Sound Effects: Insert non-speech audio in [lowercase brackets]. 3. Clean Up: Remove filler words (um, uh) and false starts. 4. Formatting: Max 42 characters per line. 5. FSU Standards: Ensure 'Florida State University', 'FSU', and campus locations like 'Landis Green' are capitalized correctly. Here is the raw transcript: [PASTE YOUR TEXT HERE]
Prompt 2: The Audio Description Scripter
Act as a Video Accessibility expert. I am providing you with the dialogue script and visual context. Write Audio Description (AD) cues. Rules: 1. Identify "Natural Pauses" where a description fits without talking over the speaker. 2. Use present tense (e.g., "The student walks..."). 3. Read aloud any important on-screen text that isn't spoken. Format as: [Time/Pause] - "AD Script" Dialogue/Context: [PASTE SCRIPT & VISUAL NOTES HERE]
Key Steps for Adding Audio Descriptions:
Adding audio descriptions for ADA Title II compliance involves creating a narration track that describes essential visual information (actions, scenes, on-screen text) during natural pauses in dialogue. Methods include creating a separate, dubbed video file, utilizing video editors (e.g., Premiere) to mix in a second audio track, or using player features on platforms like YouTube to upload descriptive tracks.
- Scripting: Write a script describing meaningful visual information, such as on-screen text, speaker changes, or crucial actions that are not explained by the main audio.
- Recording: Record the script using audio software, ensuring the voiceover does not conflict with the primary audio.
- Integrating: Combine the new, descriptive audio track with the original audio using video editing software, placing descriptions in pauses.
- Publishing Methods:
- Separate File: Create a specific "Audio Described" version of the video.
- Platform Tools: Use the "Descriptive audio" feature in YouTube Help to upload a separate audio file.
- Accessible Players: Utilize players like Able Player that support audio description tracks.
- Alternatives: If an audio track is not feasible, provide a detailed text transcript that covers both audio and visual information.