Voice and data processing

Blueworx Voice Response runs on an IBM pSeries computer that has the AIX operating system and a number of other licensed program products (LPPs) installed on it. The following types of voice applications are supported:

CCXML 1.0 with VoiceXML 2.1
VoiceXML 2.1
Blueworx Voice Response Java application programming interface
The Blueworx Voice Response proprietary state table interface

In addition, to extend the system capabilities, you can create custom servers by using the C or C++ programming languages.

The detailed information in Blueworx Voice Response for AIX: General Information and Planning should be used to determine which programming models are best suited for your system. Having made that decision, refer to Software prerequisites to determine what other LPPs you need to install before starting the installation of Blueworx Voice Response.

Speech recognition

Speech recognition by MRCP Version 1-compliant speech recognition products is supported in both the VoiceXML and Java programming environments. The runtime components of WebSphere Voice Server or other MRCP Version 1 compliant speech recognition product such as Nuance Speech Server version 5.1.2 (Recognizer 9.0.13), or Loquendo Speech Server V7 work with the speech recognition engine to perform the recognition. Recognition engines can be distributed across multiple systems so that resources can be shared, and redundancy is provided.

For detailed information about the speech recognition functions provided in WebSphere Voice Server, refer to the WebSphere Voice Server infocenter at:

http://publib.boulder.ibm.com/infocenter/pvcvoice/51x/index.jsp

Text-to-speech processing

Text-to-speech (TTS) processing by MRCP Version 1-compliant text-to-speech products is supported in both the VoiceXML and Java programming environments. The runtime components of WebSphere Voice Server or other MRCP Version 1-compliant text-to-speech product such as Nuance Speech Server version 5.1.2 (Vocalizer 5.0.2), or Loquendo Speech Server V7 work with the speech synthesis engine to perform the text-to-speech processing. TTS engines can be distributed across multiple systems so that resources can be shared, and redundancy is provided.

For detailed information about the text-to-speech functions provided in WebSphere Voice Server, refer to the WebSphere Voice Server infocenter at:

http://publib.boulder.ibm.com/infocenter/pvcvoice/51x/index.jsp

High-quality sound recording and playback

To record high-quality sound or use audio editing software on a separate personal computer.