Per request from the alpha group, the talking UI now stores the things it speaks in a temporary memory cache, allowing users to navigate word-by-word through previously spoken text.
Basics: use the arrow keys to navigate. Up arrow takes you to the beginning of the cache; down arrow to the end; left arrow to the previous word; right arrow to the next word. The words will be spoken as you reach each one. Press the Escape key to hear the word spelled out letter-by-letter. This is intended to help with spellings of unusual names or game objects.
Nuances: while navigating within the cache, new text arriving from the game will be spoken, but the contents of the cache will not be replaced. The cache will remain until you type a new command; then it's replaced with whatever's spoken next. Thinking behind this is to prevent the cache being changed while you're scrolling through it.
Caveats: it seems to work pretty well! But there are a couple of odd corners we haven't fixed yet. You may have to press the down arrow or escape key twice, since the arrows are concurrently used by the browser to move around inside the input text field. Not sure if we can finesse that, or not. But: it does seem to work!
Let us know what you think. Is this useful? Intuitive? What would you change?
This message was edited 1 time. Last update was at 2014.10.31 03:53:14