Developing X+V Applications Using the Multimodal Tools

Continued from previous page...

Getting started

The Multimodal Tools are supported on Microsoft® Windows® 2000 and require at minimum an Intel® Pentium® 500 MHz processor or equivalent.

In addition, you must have an installed version of either:

WebSphere Studio Site Developer V5.0
(refer to http://www.ibm.com/software/ad/studiositedev/)
or
WebSphere Studio Application Developer V5.0
(refer to http://www.ibm.com/software/ad/studioappdev/).

A single launcher panel simplifies the installation of the total package, including the Voice Server SDK, Multimodal Browser, and Multimodal Toolkit.

Getting up to speed quickly

The Multimodal Toolkit offers rich functionality and many resources to maximize productivity:

• Getting Started Guide (PDF format) with a basic scenario that you can re-create on your system to become familiar with the steps for creating a basic multimodal application.
• Online help (opened using Help > Help Contents > Multimodal Tools) with detailed information about creating multimodal projects, including X+V files, grammars, pronunciations, and audio files.
• Content assist with popup windows in the editors that list valid tags and elements, as well as descriptive information at the cursor position in the editor.
• Related documents including the Reusable Dialog Components guide and the Voice Server SDK VoiceXML Programmer's Guide, which provide detailed information, samples, and snippets that you can use in your applications.

Now take a look at the Multimodal Toolkit. To do this, open WebSphere Studio, and then change to the Multimodal "perspective" (or view) by selecting Window > Open Perspective > Multimodal. To see the Welcome screen, select Help > Welcome > Multimodal Tools. The figure below shows the Multimodal perspective.



The Multimodal Toolkit perspective is composed of four panes. Looking at the illustration above, you can see how the Navigator, Outline, source editor (showing the Welcome message), and Tasks panes provide a coordinated layout of the information most useful for the application developer. The panes can each be resized, dragged to another location, or closed independently, letting you customize the view.

With a few clicks, you can create the initial folder and basic files needed to start an application. Experience with HTML Web pages and basic VoiceXML will increase the speed of development. The toolkit is rich with features that simplify each step. Each editor, including the X+V editor, has customized menus, right-click (contextual) menus, and toolbar buttons. A new tab appears above the source editor for each open file. And you can use the Import wizard to add your existing files into the toolkit.

Creating an X+V application

The following flowchart illustrates the basic process of completing a multimodal application using the Multimodal Toolkit.


The flowchart shows each milestone in the development process, which might also involve multiple substeps. The details in performing each of the substeps is beyond the scope and the objective of this paper; however, the Multimodal Tools comes with a Getting Started Guide, a how-to manual for developers, which includes guided practice in the substeps not described here.

A sample application

With the flowchart in mind, let’s look at an example of an application that you might find on the Web today, shown in the figure below. In this example, you will see how to voice-enable fields in a typical transaction page that collects a customer’s billing address and credit card information.



This sample page consists of multiple components: address, city, state, zip code, phone, credit card type, credit card number, and expiration date. A complete application would consist of even more. For the purposes of this document, our example will focus just on the city field and credit-card type field.

Tip: If you are starting a multimodal application from scratch, without any existing HTML, it is always a good idea to write and test the visual portion of the application first, keeping in mind that the next task will be to determine which portions on the visual page you want to voice-enable.

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).