I'll be collecting, on this page, my material related to the VoiceXML Plugin of voice enabled Opera:
In the following list all CSS3 Speech properties supported by Multimodal Opera are given. To each property there is one testcase with the possible values. You can cycle through all testcases by means of the site navigation mechanism - the next link brings you to the next test. In each test the help link brings you directly to the relevant bit in the spec.
All testcases are written in XHTML+Voice, and are served with the correct MIME type of application/x-xhtml+voice+xml if your browser supports it.
Note for Safari users: I got reports that Safari 1.2 reproduceably crashes on the XHTML+Voice pages.
Depending on what exactly you want to do with voice, you might want to optimize the setup for your needs. I am offering some of my personal experimenting - comments and feedback are welcome!
This setup allows you to browse web documents or your mail with your eyes off the screen. It comes in two setup files - you can opt for either one, or combine them. For any command that highlights anything, the browser automatically speaks the selection, telling the user where exactly he is now.
Download the keyboard setup file.
In a similar fashion all c3n voice commands mapped to document navigation in the Browser Widget section will be enriched with a & speak selection command to tell the user where he moved to.
Download the voice setup file.