Hearing loss affects more than 28 million Americans. Captions and transcripts allow the content of web audio and video to be accessible to those who are aurally impaired, and more broadly, to those who do not have access to audio, or who are in a quiet setting. They are also useful for those not fluent in the native language of the audio or video.
Not only do captions and transcripts make audio-visual media accessible, they also increase the searchability of your media. In addition, research shows that subtitles can also increase the amount of time that a user spends watching a video by almost 40 percent. In fact, where subtitles appeared, 80 percent more people watched the entire video to its completion. View the Reel SEO article on captioning statistics.
You should begin to caption as part of your video production process. Captioning is no longer difficult to implement nor prohibitively expensive. You can hire a service to produce your captions or caption in-house. Captioning does take some adjustments in your web production process, but once you have mastered the techniques involved, captioning is straightforward.
There are many production possibilities, as this video showing someone using MacSpeech Dictate and MovCaptioner illustrates. You can quickly develop a process that works for you.
If your video is over 10 minutes in length and was not produced using a transcript, seriously consider sending it to Automatic Sync Technologies to be transcribed and captioned. (Three other transcription services with good reputations are Casting Words, 3-Play Media, and Tech Synergy.) AST will both transcribe and caption your video at $159 per media hour ($2.65 per minute), with a standard 72-hour turnaround (they are usually much quicker than that). Rush pricing is also available. They make it very easy for you to sign up and get running. Once you receive your caption file, you just need to add it to your video.
You may be aware that YouTube will automatically generate captions for your videos if you do not supply your own captions. Machine-generated captions are very poor quality (often comically so) and are not practical at this time. You should replace these default “captions” with captions that are actually useful. One redeeming quality of the machine-generated captions is that they can be used as a starting point for captioning your own videos, discussed in the next section. Time stamping the phrases, even if many or most of the words are wrong, is still a big time saver. Using an editor, you can modify machine-generated YouTube caption files, by downloading them or editing them online, and replace the initial text of the machine-generated caption with your reworked text. Note that if the sound track on the video is of poor quality, e.g., it has a lot of background noise, autocaptioning may fail.
Captioning your own videos
There are many different techniques for captioning your own videos. This section will focus on a few of the most common ones. There are four basic steps to captioning:
- Producing the transcript
- Syncing the transcript to make a caption file
- Editing the caption file
- Publishing the video and caption files
Producing a transcript
Producing the transcript is the most time-consuming and fatiguing step in the captioning process. However, if the video is of the 5–10 minute variety and you have the time and inclination (or are on a bare bones budget), it is possible to produce transcripts yourself relatively easily.
To transcribe a video yourself, type the transcript into a simple
.txt file with a the text editor of your choice while playing the video. The VLC Media Player is a free, cross-platform media player with good controls—important, because you need to stop and start the video with precise timing. Another advantage to using VLC is that it plays almost any video format. After typing in your transcript, you will want to save the transcript file to a text-only format. Specialized software, such as Express Scribe, can also make transcribing easier.
If you will be doing a large volume of transcribing, you will want to automate the transcription process as much as possible. One trick is to use dictation software to reduce the effort. There are two popular dictation software programs produced by Nuance that are invaluable in transcribing:
These programs are much more accurate than the machine transcriptions produced by YouTube. The reason for this is that before you start using the dictation software, it samples your voice while you are reading a few paragraphs to produce your own custom voice profile.
One technique is to play increments of your video, pausing the video to repeat the words with the dictation software. This will save much time and be less fatiguing typing everything. If the speakers are not talking too fast, you can save even more time by playing the video continuously and reflexively echoing the speakers words as they are said—believe it or not, this technique can be much easier than trying to hold phrases in memory and then speaking them. With either technique, it is faster to save the punctuation and corrections for a second editorial pass.
Syncing the transcript to make a caption file
Before syncing your transcript, make sure the transcript matches as closely as possible what is being said in the video—even if it is grammatically incorrect or the speaker false starts or changes direction mid-sentence. Syncing software matches the phonemes on the audio track to the words, so you want a good “sound” match, not polished editing. You will clean up any rough spots during the editing stage.
Once you have your transcript, there are two basic options for syncing your software. If your video segment is less than 10 minutes, you can upload the video to your YouTube account and use Google’s syncing tool to produce a .sbv caption file. It’s free and lets you download the resultant caption file. If you want to convert the YouTube captions to a different format, say an W3C .xml file for a Flash Player, you can use the OSU web-based caption conversion utility.
A second option is to pay Automatic Sync Technologies to do the syncing and captioning for you.
Editing the caption file
Syncing technology is pretty accurate, but you will still want to edit the caption file to correct for errors in the speaker’s delivery, bad line breaks, and other artifacts of the processing. As with the transcription process, you can use any editor that will save a document to a text file. You may wish to use a caption edting program, such as MAGpie (Windows), MovCaptioner, YouTubeCC, or CaptionTube.
For information on transcript formatting conventions, view the WGBH captioning FAQ.
Publishing the video and caption files
Just as important as captioning your videos is making sure that your video player is keyboard accessible. The two most popular video platforms on campus are YouTube and Flash. If you are publishing videos in iTunes U, you can download the WGBH document, Creating Accessible iTunes U Content (PDF, 2.8 MB).
YouTube’s player is Flash-based (although there is an HTML5 beta site) and may not be not keyboard-accessible in some browsers. Furthermore, some screen readers have difficulty accessing Flash controls and others cannot access them at all. Fortunately, Ohio State University has published an
Ohio State also has made available an accessible Flash player module based on the popular JW Player, which itself now comes with captioning and keyboard support. Either module can be imported into your web page and easily customized to support publishing your Flash video.
If you are intrepid and feel comfortable being on the bleeding edge, you can also write your own accessible HTML5 player.