The Voice eXtensible Markup Language (VoiceXML) is an XML-based, industry-standard language for creating voice applications, much as HTML is a language for developing visual applications.
VoiceXML is defined and promoted by an industry forum, the VoiceXML Forum, founded by AT&T, IBM, Lucent and Motorola, and currently supported by 575 member companies. VoiceXML was designed to create audio dialogs that feature text-to-speech, digitized as well as prerecorded audio, recognition of both spoken and dual-tone multi-frequency (DTMF) key input, recording of spoken input, telephony, and mixed-initiative conversations. Its goal is to provide voice access to Web-based content and applications. It enables the development of voice applications via the use of a familiar markup style and Web server-side logic to deliver applications over telephone lines. The resulting applications allow conversational access to Web-based data, and can also interact with existing back-end business data and
logic. VoiceXML supports dialogs that feature:
- Recognition and scoping of spoken input
- Dual-tone multi-frequency (DTMF) input
- Recording of spoken input
- Synthesized speech output
- Prerecorded digitized audio output
- Dialog flow control
- ANI and DNIS
A VoiceXML application is capable of retrieving information from a Web server and, by making use of scripts and appropriate grammars, the application can interact with the customer through spoken words. The role VoiceXML plays in accessing and presenting customer information and a parallel with HTML.