W3C Standardizes New Features for VoiceXML, Smarter Speech Recognition
The World Wide Web Consortium (W3C) published standards yesterday that enhance the capabilities and interoperability of voice browsers and speech recognition systems. The W3C Voice Browser Working Group has completed work on both VoiceXML 2.1 and Semantic Interpretation for Speech Recognition (SISR) 1.0, two critical pieces of W3C's Speech Interface Framework.
VoiceXML 2.1 extends the widely used VoiceXML 2.0 dialog language to include commonly implemented features. These features include dynamic references to grammars and scripts, detecting when during a prompt the user barges-in, and processing multiple sets of data from the server in a single access. All VoiceXML 2.0 applications will work under VoiceXML 2.1 without modification, allowing for significant interoperability.
Semantic Interpretation for Speech Recognition (SISR 1.0) enables developers to extract and translate textual representations of words recognized by a speech recognition system and structure the results into a format convenient for processing by the speech application. For example, with SISR, one could specify the instructions for converting the spoken sentence "I want to fly from Los Angeles to Seattle." to a data structure containing "departure: LAX" and "destination: SEA".
W3C's Voice Browser Working Group includes leaders in telephony and Web applications development, including Aspect Communications; Chinese Academy of Sciences; Comverse Technology; Deutsche Telecom AG; France Telecom; Genesys Telecommunications Laboratories; HP; IBM; iFLYTEK; IWA-HWG; Intervoice; Korea Association of Information and Telecommunication; Loquendo S.p.A.; Matsushita Electric Industrial Co.; MITRE Corp.; Nuance Communications; Openwave Systems; SandCherry; Toshiba Corp.; Voxeo Corp.; and Voxpilot.
The W3C Speech Interface Framework includes the completed standards VoiceXML 2.1, SISR 1.0, Speech Recognition Grammar Specification (SRGS 1.0), and Speech Synthesis Markup Language (SSML 1.0). W3C anticipates that the next pieces of the Speech Interface Framework to become W3C Recommendations are CCXML, Pronunciation Lexicon Specification (PLS 1.0) and SSML 1.1. The Voice Browser Working Group is also developing State Chart XML (SCXML) and VoiceXML 3.0.