People often include screen reader users in the much larger group of keyboard-only users. Whilst this is correct (most screen reader users don’t use a mouse), it also creates a false impression of the way screen reader users navigate content.
The aria-roledescription attribute changes the way screen readers announce the role of an element. Intended to give authors a way to provide a localised and human-readable description for a role, it has the capacity to both enhance and seriously break accessibility for screen reader users.
The accessible SVG line graphs post explains how to use ARIA table semantics to make that form of data visualisation accessible to screen readers. This article uses the same ARIA based approach to make a screen reader accessible SVG flowchart.
It is common on the web for the current thing in a collection to be highlighted visually, but providing an alternative for screen reader users has often involved something of a hack. The aria-current attribute is intended to solve this problem.
Emoji help us communicate complex ideas very easily. When used in native apps and applications, emoji are reasonably accessible to screen readers, but on the web we need to do a little more to make sure everyone can understand emoji.