· Understanding Guideline 3.1 L3 SC8
· Techniques for meeting Guideline 3.1 L3 SC8
· Benefits and Examples of Guideline 3.1 L3 SC8
8. A spoken version of text content is available.
Note: This success criterion is in DRAFT form. It is presented to the Web Content Accessibility Guidelines Working Group as a formal proposal, but it has not earned the consensus of the Working Group, and it does not appear in the current Public Working Draft at http://www.w3.org/tr/wcag20. It is presented here for discussion only. It must not be cited as a normative reference.
Text content
[WCAG definition of text goes here]
The intent of this success criterion is to ensure that spoken alternatives to text content are available.
This success criterion differs from a similar requirement at L2 SC5 in that this L3 requirement is not conditional. That is, at L2 a spoken version is required only for text that requires reading ability equivalent to that of someone who has spent at least 10 years in school. At L3 the spoken version of all content is required.
This success criterion applies when the content includes text. It does not apply to text transcripts of audio content provided to satisfy Guideline 1.1 L1 SC2 or Guideline 1.1 L3 SC1.
Some users who have difficulty sounding out (decoding) words in written text find it very helpful to hear the text read aloud. This service can now be provided easily using synthetic as well as recorded human speech. For example, there are a number of products that convert text to synthetic speech, then save the spoken version as an audio file. Cost depends in part on the quality of the voice used and whether the text is likely to change frequently.
The method below is effective for small amounts of text and for longer documents that do not change often.
1. Make a recording of someone reading the text aloud, or use a tool that converts individual documents or selected passages into synthetic speech. Choose the clearest, most attractive voice if a choice is available.
2. Save the spoken version as an audio file. Use an audio format that is widely available and supported by media players.
3. Provide a link to the audio version.
4. Identify the audio format (for example, .MP3, .WAV, .AU, etc.).
5. provide a link to a media player that supports the format.
Server-based methods may be best when pages change often or when user choice determines text content. Some server-based tools allow users to select any text they are interested in and listen to it. Typically, the user presses a button which starts the text-to-speech conversion and reads the text aloud.
· Use the Digital Talking Book format to synchronize the spoken version with the text. This can also highlight text as it is spoken, helping some users to keep track of their place in the text.
Benefits: How Guideline 3.1 L3 SC8 Helps People with Disabilities
This success criterion helps people whose disabilities make reading difficult or impossible. These include:
People with learning disabilities or cognitive limitations that impair the ability to read
People with low vision
Older persons with conditions such as macular degeneration, cataract, or glaucoma that limit reading, yet who do not use assistive technology
Learning disabilities such as dyslexia affect the ability to recognize individual words. This is called decoding. The act of decoding text word by word consumes much of the mental energy that most people are able to use for understanding what they read.
People with learning disabilities are able to understand very complex ideas and processes. People with learning disabilities have become successful scientists, engineers, artists, attorneys, and skilled professionals in many other fields. These individuals are often at a significant disadvantage when ideas and information they need are available only in complex textual presentations. Yet these same people can work as equals when ideas and information are presented in spoken form.
Example 1: A “Read aloud” button on a site that offers medical information to the public
A medical school operates a Web site that explains recent medical and scientific discoveries. The articles on the site are written for people who are not doctors. Each article uses the Dublin Core metadata specification to identify the education level of the intended audience as having completed 8 years of school and includes the Flesch Reading Ease score for the article. A button on each page says “Read this article aloud.” The user selects the button to hear the article.
There are a number of tools that convert text to synthetic speech and save the result as an audio file.
[Note: The inclusion of a product or vendor name in the list below does not constitute an endorsement by the Web Content Accessibility Guidelines Working Group or the Web Accessibility Initiative of the World Wide Web Consortium. This list is provided simply for convenience.]
· TextAloud converts text files to synthetic speech using voices created by AT&T. Retrieved 7 April 2005 from http://www.textaloud.com/.
· Text-to-Audio converts text to synthetic speech and saves the result as an audio file. Also uses AT&T voices. Retrieved 7 April 2005 from http://www.premier-programming.com/TTA/texttoaudio.htm.
· Browsealoud is a server-based solution designed specifically to support users with learning disabilities. For information, see http://www.browsealoud.com/. Retrieved 7 April 2005.
· Loquendo provides a number of server-based solutions for converting text to synthetic speech. Retrieved 7 April 2005 from http://www.loquendo.com/en/technology/TTS.htm.
· The DAISY Consortium provides information about tools for creating Digital Talking Books. Retrieved 7 April 2005 from http://www.daisy.org.
· AMIS is an OpenSource reader for Digital Talking Books (DAISY books) with multilinguial support. Retrieved 7 April 2005 from http://www.amisproject.org/index.html.