Jump to: navigation, search

Deploying SpeechMiner

This section describes the SpeechMiner system and how to configure each component for your enterprise.

SpeechMiner Components

The SpeechMiner system makes use of the following components:

  • UPlatform service—Manages all the processing tasks of SpeechMiner—fetching (in the case of Analytics mode), recognition and exploration (in the case of Analytics and Analytics & Recording UI modes), categorization compression, and indexing (in all modes).
  • Recognition engine—Nuance speech-recognition engine that transcribes call audio into text.
  • Nuance License Server— This server manages the Nuance engine. During installation, you need to either install a new instance of the Nuance License server, or provide the details of an existing server.
  • UConnector service—Retrieves interaction data (audio or text) and metadata from the recording systems that is placed in the SpeechMiner input folder.
  • Interaction Receiver—A web service which receives calls (audio and metadata) from the Genesys Interaction Recording system.
  • MS-SQL database—The SpeechMiner database stores the interaction data and the results of interaction processing.
  • Web service—Runs the SpeechMiner web-based interface that enables users to view and work with the interaction data after it has been processed.
  • ULogger—The log viewer for the SpeechMiner logs.

System Software

Users employ the following software to work with SpeechMiner:

  • SpeechMiner browser-based interface—Offers a variety of ways to access the audio of calls and the results of the interaction analysis performed by the system (when Analytics mode is in use). Users of the speech-analytics system can employ this interface to:
    • find interactions that have specific characteristics or that deal with particular topics
    • identify and listen to the parts of calls that interest them
    • audit and fine-tune SpeechMiner's call processing
    • keep track of a range of system-metrics.
  • SpeechMiner administration tool (SMART)—An application that enables users to configure the speech-analytics system to search calls for specific topics and other characteristics.
  • SMConfig—An application that is used by system administrators to configure SpeechMiner.
  • SMUpgrade—An application used to upgrade the SpeechMiner database from the previous version to current version. For additional information refer to the SpeechMiner Upgrade Guide.

Installing SpeechMiner

This section explains how to install SpeechMiner at your enterprise. This section includes pre-installation steps; setting up the system components, database, and the software steps that users employ to interact with the system. The components can be installed on a single machine, or on separate machines, as required by the particular configuration of your system.

Configuring SpeechMiner

This section explains how to configure SpeechMiner after it is installed. Most of the configuration is performed in the SMConfig application. This is a Windows application that can be installed on any machine on your network. Once it is installed, it can be used, from any machine on which it is installed, to configure the entire SpeechMiner system.

This page was last edited on December 16, 2014, at 09:33.
Comments or questions about this documentation? Contact us for support!