w3c.jpg
Think document management is tricky? Try organizing audio. Because the heavy burden of CMS wasn't already convoluted and ridden with potholes, the World Wide Web Consortium happily throws a wrench into the mix with the relaunch of Voice Browser Working Group.The Voice Browser Working Group enables users to speak and listen to Web applications. Group chairmen Jim Larson, an invited expert, and Scott McGlashan of HP lead the group in its quest to standardize languages for speech capture and production, in addition to managing dialogue between users and computers. Under charters that go as far back as 2000, the Group developed a suite of specifications for W3C's Speech Interface Framework. The most recent one, State Chart XML, passed on January 24, 2006. This charter specifies the dialog flow of multimodal apps or speech. Dialog flow remains separate from information capture and rendering. The W3C Speech Interface Framework encompasses voice dialogues, speech synthesis and recognition, and call control for voice browsers. And while all these features sound like a dream for the 'net-savvy sound technician or a prococious prank-calling hack, the Framework also serves some practical purposes, covering requirements for interactive voice response applications that include use by those with hearing or speaking impairments. For inspiration-seekers, possible uses include: * Personal information access, including calendars, lists and calorie counters * User assistance in sending or receiving voice mails and e-mails * Automated ordering services or support desks, order tracking, booking or other services * Public information access, including reports on the weather or traffic conditions, local events, breaking news, or ticker and business transaction updates. Projects on the horizon for the well-rested Group include continued work on VoiceXML 3.0, which builds advanced speech apps in a form that can be simply integrated with other W3C languages. Several versions of the work in progress are on the pipeline for publication. SSML 1.1, which manages speech synthesis, will also see a refurb, as the Group (always with a capital G) plans on adding neat little components like emotion elements. Some time ago SRGS 1.0, a speech recognition solution that focuses on context free grammar, received a full Recommendation. To build on that, the Group is considering resuming work on N-Gram (statistical) model of speech. Multiple documents are currently in the works for inclusion in the white circle of W3C Recommendations. Some, however, are not so lucky. The following are not expected to move onto the next level: * Pronunciation Lexicon Specification Version 1.0 Requirements * Speech Synthesis Markup Language Version 1.1 Requirements * CSS3 Speech Module The Working Group is chartered to hold out until January 31, 2009. Expect the first face-to-face meeting to take place in May or June of this summer. If you're a W3C member who would like to join the Voice Browser Working Group, get involved here. Otherwise, you can read more about Voice Browser.