The attached file is a very simple test of screen-reader support for CSS-generated content, defined using the following syntax:

p:before { content: "You passed, you did not " }

so that, if rendered aurally by a screen reader, the CSS-generated text ("You passed, you did not "), in conjunction with the first word of the paragraph, ("FAIL!"), informs the tester whether the CSS-generated content is available to the tester. otherwise, the aural rendering of the sole paragraph on the page will begin: "FAIL!".

Note that the expected result -- even with User Agents that support CSS generated content -- is that the attached document will fail an aural test, as the generated content is not available to the screen reader via the DOM, nor are most stand-alone screen readers CSS-aware. CSS-generated text is written directly to the screen, and hence is unavailable to the Assistive Technology.