Requirements: Text-to-speech supported

[shawnthompson]

The text-to-speech subgroup have been discussing the text-to-speech requirement in WCAG 3: Text-to-speech supported, gathering information we could find on text-to-speech in general and could not find anything specific to text-to-speech devices that would not also be found in a screen reader device.

Further reviewing WCAG 3, we would like to propose a suggestion moving Detectable text from 2.2.3 Clear meaning into 2.2.2 Text-to-speech, as we believe detectable text is the foundation of text-to-speech technology. This would replace the Requirement: Text-to-speech supported.

We were also wondering when we say “Text-to-speech” are we referring to the technology of text-to-speech or the ability to convert text-to-speech? If we are referring to the ability, perhaps we should rename the section “Text as speech” or something to avoid confusion with the term of the technology.

On the topic of text-to-speech as technology, where would user agents / text-to-speech application developers’ guidelines reside within WCAG 3?

[mraccess77]

One thing that comes to mind in support of text to speech (and not a screen reader) beyond being in text - is the ability to select text. Many folks who use that feature tend to use a pointer to select text to speak. There are some related items as well such as making sure what is spoken matches what appears visibly and not replaced by some other accessible name.

[shawnthompson]

Thanks @mraccess77 for adding that. I'm wondering if selected text be a prerequisite (insert whatever term you want to use here) for the Detectable text requirement as it would relate to other needs like converting text into plain language or translate text and not just text-to-speech.

I've been on sites with sensitive information (language exam) that would not allow you to select the text and I couldn't use my text to speech technology, even if I tried to get it to read the whole page without selecting text.