Hi.

The problem is most action browser games, whether short arcade style or full on mmorpgs like runescape or puzzle pirates  very heavily graphical.

There have been one or two attempts to build arcade style games in flash, but even there screen reader support can be interesting, one reason why the flash version of blind swordsman was better downloaded than played online.

The only browser games I know that are not text based  drive, sudosan, and These games from gamevial such as colonel zeeg and rebound.

Actually the gamevial ones are pretty cool as far as they go (I rather enjoy rebound), but obviously they're smaller arcade titles.

The closest thing we've got to something more complex, multiplayer and realtime would be either muds, or the few sound based mmorpgs we have such as survive the wild and before it's demise deathmatch, though manifestly those are client based rather than browser based.

I confess at this stage of development, where everyone has a working internet connection I do wonder what value there might be to creating browser based action audiogames, especially given that the platforms used to create them such as unity and flash have fairly dodgy relationships to screen readers, where as a custom client of course can be %100 screen reader friendly.

I suppose there is the question of cross platform availability,even then you have the screen reader issue, though I suppose there might be some cross platform potential if you could! get flash or unity to create a self voicing game and thus obviate the need for a screen reader, ---- though again good 3d sound support is in either platform I don't know.

Getting back to the question at hand, Drive, sudosan or the above mentioned gamevial games is probably your lot thus far, though on the plus side if you haven't played the gamevial titles I'd recommend giving them a go.

HmmmI'd forgot about html5, though again why someone should now create an html5 game online rather than an offline client I'm not sure, unless cross platform compatibility was the basic idea, though you would have the advantage in html5 of not being reliant on dodgy flash support in screen readers, eg, you could just label your "start game button" and be confident that people with whatever screen reader could find and click it, though manifestly this might not be the case for real time text within the game itself, plus of course with html5 you'd be dependent on the browser for your in game speed especially across multiple games, and it'd be fairly dire if one player got the advantage because the other was stuck with some dodgy net interference.
this already is a problem in playing some mainstream titles online even across things like playstation network (I recall my brother cursing net lag when playing street fighter 4).

The disadvantage on that list is the need to use espeak or similar, since generally they don't sound pretty and we get enough complaints when someone makes a game that runs just with sapi, (especially since these days even the default windows sapi voices are light years beyond old Microsoft sam).

As for "developers to help", well I can see the logic of having a very well known and universal audio development language from a programming and theoretical perspective, unfortunately practically people are just not that interested in making audiogames, and the majority of those who are use either python, bgt or have their own more idiosyncratic preferences in languages and systems they are already familiar with, like Aprone and his vb code.

In the end the proof is in the pudding, and it'll take someone trying to build something to see if it is actually worth while I think, rather than theoretical speculation, after all I recall a time when everyone was sure that Java and the swing access bridge was the next big thing.