Strategies
Related Guidelines:1 WCAG 1.2.1, WCAG 1.2.2, WCAG 1.2.3, WCAG 1.2.5, WCAG 1.3.3, WCAG 1.4.1, WCAG 1.4.2, WCAG 1.4.3, WCAG 1.4.11, WCAG 2.1.1, WCAG 2.1.2, WCAG 2.3.1
Many courses contain some form of recorded video content, such as lecture recordings in Zoom, short-form social media videos, recorded slide presentations, etc. Because learners have diverse needs, it’s important to keep accessibility in mind when incorporating recorded video content into your course.
Whenever possible, provide accurate captions and a transcript for all videos, unless the video has no content-related audio (e.g., no narration, voice-over, dialogue, etc). In that case, be sure to include an audio description track2 for students who are blind or have low vision, to help them fully access your course content.
Ideally, videos should have both captions and a transcript because the primary audience and purpose for each option are different. Below, we outline a few key reasons to use both and also offer definitions and examples for each.
Captions
Captions3 are helpful to many individuals, but they are especially necessary for those who benefit from auditory accommodations.
When it comes to generating captions, here are a few options you can try:
- Review and fix the video’s automatically generated captions for accuracy.
- Manually transcribe and synchronize the captions for the video yourself.
- Pay a professional service to generate the captions for the video.
Most instructors choose to fix the automatically generated captions, which works well in many cases. However, there are a few situations where the quality of the automatically generated captions may be too poor to be useful, and trying to fix them will not be the fastest solution. An example of this is when there are multiple speakers in the video, or the video has poor audio quality.
Transcripts
Transcripts4 are especially necessary for those who need some degree of both visual and auditory accommodations and may be used in sync with an assistive device such as a braille display.
When it comes to generating transcripts, here are a few options you can try:
- Use a platform and/or video player that automatically generates a transcript when the video already has captions, or that makes it possible to export a transcript from a captions file.
- If you recorded the video yourself with a script, you can modify that script into a transcript.
- Listen to the video and transcribe the audio by hand. This option can be time-consuming, so it’s important to plan ahead if selecting this option.
Please note that not all video platforms have the feature to automatically generate a transcript from a captions file. However, in that case, you don’t necessarily have to sit down and transcribe the whole video, line-by-line, with no support. There are online tools that can help make transcribing a video by hand much easier. If you have questions or need further support with manually transcribing a video, you can schedule a consultation with us at cornellCTI@cornell.edu, and we’ll be happy to recommend online tools to help streamline this process and save time.
When captions and a transcript are not enough
There are some situations where captions and a transcript are not enough to make a video fully accessible. Here are some examples:
There’s no content-related narration, voice-over, or dialogue
Some videos may have no audio, or their audio doesn’t describe what is happening in the video. Examples of this are videos that only have sound effects or background music that is not the main focus of the video (Note: music videos are an exception in some cases). Closed captions and a transcript are of little use in this situation. This is not an issue in most courses, but you do occasionally see this with animated simulation videos, montages, screencasts, films, or short-form videos on social media, etc.
If possible, try to avoid using these types of videos. Videos that do not have descriptive enough audio require retrofitting with audio descriptions — in other words, creating a separate audio track for learners who cannot see the videos. Audio descriptions can be challenging to create, so a simpler alternative is to plan in the early stages of the video creation process to include descriptive audio in the form of an instructor talking, or with dialogue, a voice-over, narration, etc.
You’re not using a Cornell-supported video platform
Believe it or not, the video player itself can also be a barrier to some learners. Not all learners can use a mouse to play, stop, pause a video, or enable closed captions, etc. Some learners can only use the keyboard, possibly in combination with assistive devices, to access the content. Unfortunately, not all video players, apps, or websites have accessible code that allows users to use keyboard shortcuts. This is why it’s important to use a video platform that is keyboard navigable.
For this reason, we strongly encourage using a Cornell-supported video platform, as they likely have already gone through a basic accessibility review for this issue. The only exception to this recommendation is if you know for certain that a reliable source conducted an accessibility review for that video platform and thus, you can definitively confirm that it follows key accessibility guidelines, such as keyboard accessibility.
Additional Strategies
Additional accessibility strategies that apply to most video content:
Graphics
If the video contains graphics, demonstrations, labs, scenes, or any other visuals that convey information or a story, try to ensure that learners can glean that information or story using the audio alone.
If possible, avoid putting important text or visuals at the very bottom of the video. This is because captions are often found at the bottom of the video player by default and, as a result, may block some of your content. Some video players allow users to move the captions around the screen, but not all learners will be able to use this feature and/or be aware that it exists.
Use high-quality graphics whenever possible. Low-quality graphics are difficult for many learners to see and interpret.
Video Player
If possible, use a platform and/or video player that is supported by Cornell (e.g., Panopto, Kaltura, Zoom). These are supported by Cornell staff and are more likely to have basic accessibility features. For example, they are more likely to be keyboard navigable5.
Some video players have a setting that enables the video to play automatically when a person loads the page. Try to disable this setting because the audio in the video can overlap with that of a screen reader6, interfering with student comprehension of your content.
Colors
Try not to rely solely on color or visual appearance when giving instructions. For example, when demonstrating how to do a task using specific software, instead of saying “Click the red button”, you can say “Click the submit button” or even “Click the red submit button.” This helps students who are color blind or who have low vision access the material.
When choosing graphics for the video or designing slides for a recording, we encourage you to use a color contrast checker tool whenever possible. For more on this, review Color Contrast.
Animations
Whenever possible, try to avoid using videos with fast flickering lights, effects, and/or animations. For some learners, fast lights, animations, and effects (depending on speed, hue, brightness, and other factors) can trigger a seizure. In addition, flashing lights can be distracting and/or cause headaches. If it is necessary to include a video in the course with these effects, we recommend including a warning at the beginning of the video or in a prominent spot on the same page where the video is hosted.
Text
Remember to explain all acronyms, abbreviations, jargon, slang, and metaphors in the video itself OR adjacent to the video on the page, or in a linked document.
Try to use large font sizes and unembellished typefaces.
If preparing slides for a recording, try to limit the amount of text per slide to a few short phrases. The more text you add to the slide, the smaller the font size becomes and, as a result, the text becomes more difficult to read. See the Slides section for more information.
Tips
Try short-form videos
Keep in mind that short-form videos (5 mins or less) that are focused on a single topic are often more effective learning tools than long lecture recordings. Not only are students more likely to watch the entire video, but they are also more likely to remember important points and rewatch the video for studying purposes. In addition, using short-form videos will reduce the time needed to caption and transcribe.
Remember that fixing the auto-captions is not always the easiest solution
There are pros and cons to each of these strategies we’ve discussed. While fixing automatically generated captions seems like it would always be the easier option, you may be surprised to find that this is not the case universally. The accuracy of automatically generated captions varies considerably based on many factors. For example, they may be very inaccurate if your video has considerable subject-matter-specific terms or jargon, was recorded with a low-quality microphone, or has multiple people speaking at once. In some cases, bad automatically generated captions can take longer to fix than writing the captions by hand. We recommend keeping an open mind when it comes to captioning options, as this is something you will have to judge on a case-by-case basis.
Plan ahead before moving a video from one website/platform/tool to another
If you own a video that already has accurate closed captions, this is something to keep in mind if you plan to move that video to another website, platform, or tool in the future.
The vast majority of captions on the web are closed captions, which means they can be turned on and off with a CC button in the video player. This is possible because the captions are just an overlay on the video and are not permanently burned into that video file. Closed captions are uploaded to the video player separately from the video in the form of a captions file that contains the text and timestamps.
If you download the video from one platform, such as YouTube, and upload the video to a different platform, such as Panopto or another Cornell-supported tool, the captions will not automatically transfer to the new platform along with the video itself. In most cases, you will need to download the caption file from the original platform and upload it to the new platform separately from the video.
Note: Different websites, platforms, and tools may require a different file format for captions. Be sure to plan ahead and determine which file formats you will need before moving the video: 3PlayMedia Closed Caption Formats.
Try informal testing to check for keyboard accessibility
Throughout this guide, we reference keyboard navigability, which is a way for learners who struggle with using a mouse and rely on keyboard controls to navigate their courses. Other assistive technologies, including screen readers and voice recognition, also rely on this form of accessibility. You may be surprised to know that you can step into your students’ shoes and do a quick, informal test to determine whether your video player is keyboard navigable. This technique is not infallible, but it can alert you when further testing is urgent. Here’s how to check:
- You should be able to navigate to each button in the video player by repeatedly pressing the TAB key. The buttons on the video player should also highlight as you TAB to them.
- Once you get to the play button on the video player, press ENTER/RETURN to activate the button.
- If you are unable to navigate to the play button via the TAB key and play the video with ENTER/RETURN, the video player may not be keyboard accessible.
- Try this with the other buttons on the video player as well; e.g., the CC button.7
Specific Tools
Kaltura
Captions
Option 1: Fix the Automatically Generated Captions
All videos uploaded to Kaltura automatically generate captions; however, it may take time for the CC button to appear in the video player. How long this takes depends on the length of the video. For example, long lecture recordings may take several hours or days to generate, whereas short-form videos may take seconds or minutes.
Once the captions are ready, they will appear in the video player automatically in the form of a CC button. (The instructor does not need to enable them in Kaltura). You can then fix the captions for accuracy. For information on how to edit automatically generated captions in Kaltura, see Editing Automatic Captions in Kaltura.
Option 2: Manually Transcribe and Upload the Captions File into Kaltura
Alternatively, you can transcribe the video by hand. There are two ways to do this in Kaltura.
- You can transcribe the video using an external tool, save your work as a captions file, and then upload that file to Kaltura. For help with this, review Upload and Manage Captions.
- You can transcribe the video by hand directly in Kaltura. This is similar to editing the automatically generated captions, but you are starting from a blank slate. For more, review Managing Captions in Kaltura in Canvas.
There are many online tools, both free and for a fee, that you can use to manually transcribe a video. If you are interested in this option and need a recommendation, schedule a consultation with us at cornellCTI@cornell.edu.
Option 3: Professional Service
There are paid professional services that work for videos hosted in Cornell-supported platforms such as Kaltura. If you are interested in this option and need a recommendation, we invite you to schedule a consultation with us at cornellCTI@cornell.edu.
Transcript
Kaltura makes it very easy to generate a transcript! The platform will automatically generate and display a transcript below the video player as long as you’ve taken steps to enable and/or create captions.
Panopto
Captions
Option 1: Enable and Fix the Automatically Generated Captions
Videos in Panopto have automatically generated captions, but they need to be enabled for you and your learners to see the CC button in the video player. You can then fix the captions for accuracy. For information on how to enable and edit automatically generated captions in Panopto, see Editing Automatic Captions in Panopto.
Option 2: Manually Transcribe and Upload the Captions File into Panopto
Alternatively, you can transcribe the video by hand, save it as a captions file, and then upload that file to Panopto. For information on how to upload a captions file to Panopto, see How to Upload a Caption File for a Video.
There are many online tools, both free and for a fee, that you can use to manually transcribe a video. If you are interested in this option and need a recommendation, we invite you to schedule a consultation with us at cornellCTI@cornell.edu.
Option 3: Professional Service
If you have a funding source, there are paid professional services that work for videos hosted in Cornell-supported platforms, such as Panopto. If you are interested in this option and need a recommendation, we invite you to schedule a consultation with us at cornellCTI@cornell.edu.
Transcript
Panopto does not give you an easy option to export captions as a transcript without timestamps, but you can download the captions file and then either manually remove the timestamps or use an online tool to do so. For information on how to download a captions file from Panopto, see How to Download Captions?
Note that using only the captions file without removing the timestamps is not an ideal long-term solution in many cases. When used in its intended form as a caption file on a video, the timestamps tell the video player when to display each sentence so that the words are in sync with the video, but the timestamps themselves are not read aloud. However, a transcript is different; it may be read by a person or an assistive device that will actually vocalize the timestamps to the learners. These timestamps are long, frequent, and sometimes cut-off sentences midway; it would be a distracting and confusing listening experience for the learners in this case.
Zoom
Captions
Option 1: Fix the Automatically Generated Captions
If enabled in your settings, Zoom will automatically generate captions for your Zoom cloud recordings. For instructions on how to do this, review Enabling the Automatically Generated Captions in Zoom.
How long it will take the tool to generate the captions in Zoom depends on the length of the recording. The longer the recording, the longer it will take to auto-generate captions. If you’d like to take this option, you don’t have to wait for this request to process. You can request to receive an email when the captions are done processing.
Once the captions are ready, you can then fix the text for accuracy: Audio Transcription for Cloud Recordings.
Option 2: Manually Transcribe and Upload the Captions File
Unfortunately, it is a complicated process to upload an external caption file directly to a Zoom cloud recording. It’s an option, but note that you will have to download the video from Zoom to your computer, caption it using a different tool, and then re-upload the video and caption file to a different platform, such as Panopto. If you are interested in this option and need guidance, please reach out and schedule a consultation with us at cornellCTI@cornell.edu.
Transcript
For cloud recordings, Zoom will generate and display a transcript next to the recording as long as it has captions. For more, review Audio Transcription for Cloud Recordings.
Context
Why are these strategies important?
One of the goals of accessibility is ensuring the content and intent of the video are perceivable and functional to the largest number of learners. Learning environments are diverse spaces, and as you likely know, many students have many different needs that must be addressed before they can fully participate in the course. When everyone has equitable access to learning, everyone wins.
On that note, keep in mind that if a learner cannot see the video or has low vision, they may rely on the audio to learn your course content. If they cannot hear the video, they may rely on what they can see. If they have limited ability to see and hear the video, they may instead rely on touch (e.g., braille display). If they have limited or no use of their hands or have low dexterity, they may rely on the use of assistive devices and the keyboard to play and stop the video player.
In addition, for some learners, it can be easier to understand and/or process the information when that information is presented in both text and audio formats. English language learners are an example of this type of student.
While there are other scenarios where someone may need to interact with the video in a certain way, we offer these examples to demonstrate why it is important to convey the content and intent of the video using many different methods.
How descriptive does the audio need to be?
How do you know if the audio is descriptive enough? In the case of a course, the instructor as the subject matter expert would know more than anyone how descriptive the audio needs to be, as it depends on the purpose and topic of the video in that course. To help determine this, we recommend reviewing the learning outcomes for that lesson or developing them if you don’t have any yet.
Ask yourself, “What should the students be able to do or know after watching that video? What information will they need to complete the course assignments?” Performing this exercise can help zero in on what information is important in that video and how descriptive the audio needs to be.
Is a written description enough if a video doesn’t have audio?
While a written description on the page may be used as a temporary aid until descriptive audio can be generated, please treat this as a temporary and last-resort option; it should not be used as a long-term solution in most situations.
How accurate do the captions need to be?
The ultimate goal is for the captions to be as close to 100% accurate as possible in all videos. Please note that it is understandable and quite normal if this takes multiple semesters to fix the captions for all the videos, especially if you have a large amount of video content in your course(s) and little support. However, what is important is that you plan to work towards that goal and make an honest effort to reach it. The key is to remember that perfection is an unrealistic expectation to have and what is important is that steps are being taken to learn about accessibility and create a more accessible course, this is what is important.