Volume 3, Issue 4 - July/August 2003
 
   
   
 

The VoiceGenie VoiceXML Platform Solution

By Rob Marchand

Introduction

VoiceGenie Technologies provides the industry-leading VoiceXML platform, the VoiceGenie VoiceXML Gateway. This platform can be deployed as Customer Premise Equipment (CPE), or within the carrier network, and is capable of hosting a wide variety of solutions, including:

  • Voice Portal Services;
  • Directory Assistance;
  • Customer Care;
  • Vertical Applications;
  • Targeted enterprise solutions;
  • Computer Telephony Integration solutions;
  • In-network Media Server;

    • VoiceGenie runs on highly reliable commodity PC-class hardware, on platforms suitable for installation in a standard data center environment, and can be purchased as a software-only distribution, or as a completely integrated hardware and software package. Although the major visible feature is VoiceXML support, the platform also includes many management and integration features allowing deployment in both enterprise and carrier environments.

    Platform Features

    The VoiceGenie platform includes the following major features.

  • Complete support for VoiceXML 2.0;
  • Advanced Call Control Features;
  • Rich TTS and ASR engine support;
  • Platform Availability and Performance;
  • Powerful Management Tools;

    • VoiceXML 2.0

    The VoiceGenie platform provides complete support for all features of VoiceXML 2.0, including:

  • Speech Recognition (ASR);
  • Speech Synthesis (TTS)
  • DTMF Detection (supporting DTMF-only deployments);
  • Audio recording and audio playback;
  • Embedded ECMAScript support;


    • VoiceGenie provides many useful extensions in the area of call control and user interface management. These extensions are not required to develop advanced solutions, but can be used to meet particular application requirements. For example, full call recording allows collection of the complete telephone call, or portions of the call, for analysis or verification purposes.

    VoiceGenie also supports an external event injection API, suitable for synchronization of VoiceXML sessions with external applications.

    Advanced Call Control

    The VoiceGenie platform provides full support for VoiceXML 2.0 call control, and supports a rich selection of PSTN environments, including ISDN, Robbed-bit/CAS, VoIP (SIP and H.323), and custom call control solutions (via an Open API) such as those encountered in Directory Assistance environments. Note also that PSTN and SIP can coexist on the same platform.

    However, beyond this, VoiceGenie provides ‘under-the-hood’ support for advanced call control, including many call release technologies (hook-flash, TBCT, ECT, SIP REFER and others), and access to data provided by the underlying network as well. Where supported by the underlying telephony hardware (currently VoiceGenie supports Dialogic and Brooktrout), advanced call progress and analysis functions are also available, allowing detection of answering machines, fax, and other devices.

    VoiceGenie provides a set of six call control tags that can be used within standard VoiceXML. These allow dynamic management of audio channels, messaging between interpreter instances, and other advanced functions, and can be used to build enhanced services such as whisper call transfer.

    VoiceGenie also supports outbound calling, from within a VoiceXML session, or as triggered by an external interface.

    Rich TTS and ASR Engine Support

    VoiceGenie supports a wide range of TTS and ASR engines, allowing the customer a range of choice of speech engine vendor. Currently supported TTS engines include:

    • SpeechWorks Speechify
    • Rhetorical rVoice;
    • AT&T Natural Voices;
    • SVOX;
    • MRCP based engines (such as Scansoft and Loquendo);

    Supported ASR engines include:

    • SpeechWorks OSR
    • Nuance 8;
    • BBN Hark;
    • Phonetic Systems;
    • Telisma;
    • AT&T Watson;
    • MRCP based engines;

    VoiceGenie allows the use of multiple engines by an application. These can be deployed on-board or as distributed components (client-server). The VoiceGenie integration of each engine exposes the particular features of the engine to the VoiceXML application where appropriate.:


    Continued...

    back to the top

     

    Copyright © 2001-2003 VoiceXML Forum. All rights reserved.
    The VoiceXML Forum is a program of the
    IEEE Industry Standards and Technology Organization (IEEE-ISTO).