Skills for Access

Author

Learning Development and Media Unit, Skills for Access Project, the University of Sheffield

Key technology

MAGPie (Media Access Generator); video; QuickTime; RealPlayer; Windows Media.

Specific Issues/Key Terms

Captions; transcripts.

Introduction

When creating the Skills for Access web site, our task was to produce a highly accessible website that was also rich in multimedia examples. We wanted to make use of video interviews and examples to add an element of personal authenticity and to provide respite from "wall-to-wall" text!

As a unit we have rarely captioned our video material, although there have been some exceptions to this when clients had a clear need for them to be done. Over the past few years we have begun to provide transcripts (frequently printable), but this has been a new departure.

Project Aim

To make video clips optimally accessible to viewers with hearing impairment. To provide an exemplar of 'best accessibility practice' by providing captioned video clips as well as text transcripts of all video clips.

Technology Used

We chose to use MAGPie to create the captions, partly because it is freely available and widely used, and partly because it is relatively straightforward.

Early in the project, we chose to make the video clips available in three formats - Quicktime, RealPlayer and Windows Media, to optimise accessibility of our material to the three most commonly used media players. This meant that each caption then had to be applied to each video format, checked and amended as necessary. The person doing the captioning work spent a lot of time getting MAGPie to work with all three video formats as different media players needed to be installed on the PC to enable MAGPie to allow the operator to playback video within the MAGPie programme. The GRiNS player application must be installed if you wish to caption media types supported by RealPlayer.

Accessibility Design Objectives

To provide a means of allowing people with hearing impairment or no sound available on their computer to access the audio component of the video clips in the Skills for Access website.

Project success

Whether or not this will be deemed to have been successful will be decided by you, the user. No mention has yet been made of the transcript production. We started with a standard transcript, produced at the going commercial rate, which then had to be checked for technical language. In certain cases, they also needed quite a lot of sub-editing - what is acceptable to the ear as a figure of speech often reads incomprehensibly on the page. There is little point in providing an accessibility aid that is hard to read.

Reflections

The skills required to caption are linguistic/production based and technical. The transcript has to be broken down into segments that both make sense and are readable on one screen. This will depend on how quickly or slowly the speaker delivers their words, which can also impinge on readability so as a last resort, some text may have to be cut (carefully, making sure that the meaning of what is said is not diminished by this editorial process).

Our captioner also found that the less 'work' done by MAGPie itself, the better - it worked best if the line breaks were already done, in Word, and the file saved as a plain text document before importing into MAGPie. She also found that the video didn't play back properly within MAGPie which made marking-up the cue points tricky as she had to rely solely on audio cues (again this is due to player versions/compatability with MAGPie). This works fine for interviews, such as the interview video clips we were working on, but would have been very tricky if we had been trying to caption something where the visual element/caption delivery required very precise timing. We also found that none of the captions played back the same way on any of the PCs in our unit, even though we are all using the same browser, and using broadly similar machines!

The following gives an idea of the time taken to alter the transcripts - i.e. re-formatting and adding line breaks - and then create captions for the video clips using Magpie. We captioned nine interview clips with a total duration of approx 15 minutes. We then applied the captions to all three video formats. This took three and a half hours, resulting in approximately 45 minutes of captioned video for the web. The interviewee did speak very slowly and concisely, making the placing of captions relatively easy when having rely on audio playback for cues.

Our experiences are reflected by material from the University of North Carolina on captioning with MAGpie:

"Learning MAGpie might take a couple of hours. The program has a couple of bugs and problems, but they are listed below which will hopefully allow you to move on with little frustration. The captioning part is the most time consuming. If you outsource your transcriptions, the time it takes to caption obviously will be cut down. Without a transcript, a 5-minute video takes approximately 2 hours to caption. With a transcript, the time is cut almost in half to approximately 1.25 hours. The whole captioning process will speed up with practice."

It seems to us, therefore, that the pivotal question that runs through the whole implementation of accessible multimedia issue is time. Where do you find the time to do it all? Of course 'time' could be replaced by 'money' since in many respects they are the same thing. There is no 'reasonable' reason for not captioning video clips from a 'widening access' point of view, but it took a number of skilled workers many tens of hours to do this site.

Where will you find the money to pay someone to do it, or what will you sacrifice doing in order to a) learn the skill yourself and then b) actually produce the captioned clips? If you don't have the resource to do the captioning or transcribing, there is a choice - provide video clips without captions and render the resource inaccessible/less accessible to those with hearing impairment or no sound on their machines, or decide not to produce video clips at all, thus denying everyone the advantage of rich media. In the end, it comes down to the 'reasonable adjustment' argument. Is there another means of providing a reasonable alternative experience for people who cannot hear the video? If not, it begs the question: should you be using video clips at all?