Voice Browser Technologies
The W3C stardards of VoiceXML and ccXML create a simple to use framework to 'voice enable' enterprise applications in the cloud or in your network. Both are simple mark-up languages and use web scripting and application technologies like Java, ASP / PHP to create 'open' interfaces with other web technologies such as SugarCRM and Salesforce.
What is VoiceXML?
VoiceXML (Voice eXtensible Markup Language) is a scripting language for writing voice enabled web services and applications. VoiceXML is the 'HTML' for telephony based speech applications, hiding the complexities of the telephony platform from developers, providing an easy way of developing feature rich and media rich speech applications.
95% IVR systems shipped in 2013 will be VoiceXML (today about 65-75%).
Top 3 applications are:
Incoming call handling / routing in Contact Center Cloud
Self-service transactions interfacing with Enterprise Cloud applications
Cloud telephony for web based outbound calling (collections, notifications, CRM)
Phonologies InterpreXer™ fully implements the VoiceXML 2.1 specification.
The ccXML Standard:
ccXML, or the call control eXtensible Markup Language, as the name suggests, is the W3C standard markup language, that provides rich telephony call-control capabilities to an interactive speech applications. It controls how phone calls are placed, answered, transferred, conferenced, and more. Ideally, ccXML works hand-in-hand with VoiceXML to provide a 100% XML based solution. However, ccXML could also be used with other dialog systems such as a traditional IVR (Interactive Voice Response) platforms created before VoiceXML was available.
The Voice Browser Working Group of the W3C has published a formal Recommendation of CCXML Version 1.0 in July 2011.
Open Source Oktopous™ ccXML Browser is the worlds most widely adopted ccXML browser.
Supporting Standards: MRCP
MRCPv2 (Media Resource Control Protocol) is a protocol for IVR, PBX, Media Servers, etc. to communicate with speech recognition and speech synthesis engines. Typically speech recognition and speech synthesis engines run on one server or in the cloud, and client servers (IVR, PBX, Media Servers, etc) can send MRCP messages over the Internet to communicate. MRCPv2 is modeled along the lines of SIP/VoIP, separating media channel for audio data streams (RTP) and control channel (SIP). Phonologies InterpreXer™ readily supports MRCPv2.