Wednesday, April 19th, 2006

ajaCSS: Accessible Ajax

Category: Accessibility

Bhasker V K has come up with ajaCSS which came from early work on:

  • Speech recognition with AJAX (watch the video demo )
  • ajax + css for low resolution displays (watch the video demo )
  • ajax + css for pattern matching / hand writing recognition (watch the video demo )

Posted by Dion Almaer at 8:52 am

3.3 rating from 45 votes


Comments feed TrackBack URI

Does anyone have any AJACSS resources. I quickly googled the idea and couldn’t find anything good and they didn’t post the slides for that online.

Comment by Brian — April 19, 2006

Those crazy guys from India

I believe I have a lot of respect for other people. And I try to encourage instead of make fun of someone. Unfortunatly, today, I read some of the wierdest stuff in a long time. From I found a news item with a few links to three projects do…

Trackback by Hallvar Helleseth — April 19, 2006

What are you talking about?

Comment by Brian — April 19, 2006

Did you read my blog post or just the excerpt here? Click my name to read it..

Comment by Hallvar Helleseth — April 19, 2006

You may be interested by a Graffiti pen recognition algorithm that I implemented in javascript. The performance is not too great though ;-)

Comment by Julien Couvreur — April 19, 2006

On plus side it doesn’t suck anymore than any other speech application.

Comment by Neuromancer — April 19, 2006

Isn’t that speech generation and not speech recognition?

Comment by J Perry — April 20, 2006

I don’t think anyone came up with an explanation of what AJACSS is or where there is anymore info about it…disappointing

Comment by Brian — April 20, 2006

AJACSS is nothing but a term made up by a confused programmer. But he explains that it is the use of both AJAX and CSS.. which is valid, but in the context, it’s just wrong. But whatever gets you going..

Comment by Hallvar Helleseth — April 20, 2006

hey ppl,
im the ‘crazy dude ‘ from india as Hallvar likes to call it .

i apologize for not giving any explanations for them videos. It was what i demoed at barcamp chennai .

Now for the speech recognition demo ,it wasnt meant for any commercial release ,only for behind the scenes implementation techniques that i used which obvisouly encapsulate what i infact did perfectly.

First up , you need to understand salt ,its compliance, the server and client side working communication that hapens before speech recognition ,as well as traditional content parsing models.
Firstly i use a combination of SALT, a lot of javascript to do two things associated with grammars (they make up the vocabulary of what to recognize,and what to do after that ) .SO while most salt recognition app’s use static gramamar(loadded during designtime) i make my grammar loaded dynamically from xml files. After that its pretty much normal speech recognition until the part where i ask to read news or files. Here i reduce the page transition time ,by using the xmlhttpobject( yup, ajax) to retrieve the files(responseText) and news (through Xmlresponse) .which means that unlike normal speech apps i do the following in a SINGLE page load :
1.load my grammar dynamicically
2.recognise what the user spoke
3. and then retirve the news/file

1,2& 3 in the same page load ! Thats the first project .

As for ajaCSS, bad explnantion again .im new to being reviewd by developers from across the world 8 )mbut im starting to admire it .ill try to make it up with the following links which i hope can throw some light on what i intended on doing .

What the links above sum up (but where the videos fail to explain) is to say that ajaCSS is meant as a prototype for being able to capture the cursor movements as a distinct reproducable xml object without any video/OCR utility involved. Effectively what this does is convert ‘x’ seconds of low resolution animation into xml ,you can play it back ,etc .

Im still developing the reader part (im done with the part where they convert what u doodle for 5 seconds into a structured xml file ) ,so thanks a bunch for the feedback …

Keep Clicking,

Comment by Bhasker V K — April 20, 2006

“in a SINGLE page load :
1.load my grammar dynamicically
2.recognise what the user spoke
3. and then retirve the news/file”

Congratulations. You’ve murdered the English language. Allow me a moment to retirve your prise.

Comment by J Perry — April 20, 2006

ha ha . 8 )

Yes,Perry it might seem very funny indeed . These are terms that you might be familiar with if you deal with speech recognition, and related programming languages in particular .

And if youre still confued ,maybe its more time to do a few google queries or read my blog more often .

Keep Clicking,

Comment by Bhasker V K — April 20, 2006

I don’t know what you meant by “confued”. Did you mean I’m “confused” or that I’m “kungfu’d”? I can assure you that I am neither, Grasshopper.

Comment by J Perry — April 20, 2006

[…] ajaCSS looks interesting. […]

Pingback by Pig Pen - Web Standards Compliant Web Design Blog » Blog Archive » ajaCSS — April 21, 2006

Check out for official news ,version updates ,etc for ajaCSS based experiments at and my blog .

Keep Clicking,

Comment by Bhasker V K — April 23, 2006

AJAX + CSS = AJXAC (Asynchronous Javascript, Xml and CSS)

Comment by Hmm — April 25, 2006

Interesting project, although I am curious how Bosky manages to a) Record audio, b) stream the audio to where it’s being ‘recognised’, and c) end-point the audio, all using AJAX?
Does anyone know how this is done?

Comment by AdamM — October 8, 2006

Leave a comment

You must be logged in to post a comment.