This call flow example shows how the Experience Portal system interacts with other systems to handle an automated telephone transaction.
A caller from the Public Switched Telephone Network (PSTN) dials a telephone number.
The PSTN routes the call to the Private Branch Exchange (PBX) associated with that number.
Using Voice over IP (VoIP), the PBX breaks the voice data into packets and sends them over the LAN to a Media Processing Platform (MPP) server in the Experience Portal system.
The MPP server looks at the Dialed Number Identification Service (DNIS) for the incoming call and uses the configuration information downloaded from the EPM server to match the number to a speech application that has been added to Experience Portal.
The MPP starts an Avaya Voice Browser session and passes it the Universal Resource Indicator (URI) specified for the selected speech application.
The Avaya Voice Browser contacts the application server and passes it the URI.
The application server returns a VoiceXML page to the Avaya Voice Browser.
Based on instructions on the VoiceXML page, the MPP uses prerecorded audio files, Text-to-Speech (TTS), or both to play a prompt to start interaction with the caller. For TTS, the MPP establishes a connection to a TTS server and the ASCII text in the speech application is forwarded for processing. The TTS server renders the text as audio output in the form of synthesized speech which the MPP then plays for the caller.
Note:
This connection requires one TTS license, which can be released as soon as processing is complete.
If the caller responds by:
Entering Dual-tone multi-frequency (DTMF) digits, the results can be processed locally by the MPP or passed to the ASR server for remote processing. The administrator selects local or remote DTMF processing when the application is configured. The digits entered are then returned to the application for further action.
Speaking, the MPP establishes a connection to an Automatic Speech Recognition (ASR) server and sends the caller's recorded voice response to the ASR server for processing. The ASR server then returns the results to the application for further action.
Note:
This connection requires one ASR license, which is not released until the entire call is complete.
If errors are encountered during the call, how these errors are handled depend on the type of grammar used by the application. If the application grammar is:
Dynamic, or In-line, the speech server gets the grammar directly from Experience Portal and any error messages are passed back to Experience Portal.
External or static, the speech server asks for the grammar using the URL specified in the application. If the URL points to an application server, the speech server interacts directly with that application server. Because Experience Portal is not involved in this communication, any error messages passed back by the application server may not be passed back to Experience Portal.
The application terminates the call when it finishes execution or when the caller hangs up.
When the call ends, the PSTN clears the call from the PBX and releases the ASR license if one was required.
