2000 Conference Proceedings

Go to previous article 
Go to next article 
Return to 2000 Table of Contents


WINDOWS-BASED SCREEN READERS: TOOLS FOR BEGINNERS AND POWER USERS

Jay Leventhal and Crista Earl
American Foundation for the Blind
11 Penn Plaza, Suite 300
New York, NY 10001
jaylev@afb.net 
crista@afb.net 

Introduction

This session will provide an in-depth analysis of Windows access with screen readers. We will review the basic tasks that a good screen reader must perform; misconceptions and realities about how to choose a screen reader; and profiles of the screen readers currently on the market.

We will then focus on our latest evaluations of Windows screen readers. This will include each screen reader's performance with a variety of Windows applications. We start with widely used programs such as Word, WordPerfect and Internet Explorer, applications on which screen reader manufacturers focus and create configuration files to maximize their products' performance. We then examine each screen reader's performance with applications that are used by fewer people including spreadsheets, databases, and programming languages. Though no single one of these applications is widely used, being able to access them with a screen reader can make the difference in getting and advancing in a job. Throughout, we emphasize that the user must know his or her screen reader.

Overview

How speech programs work

The graphical and visual nature of the Windows operating environment makes it necessary for a screen reader to do more than simply lift material from the screen and send it to the synthesizer. Its functions can be divided into five categories: 1. Identifying and reading text and graphics. Once text has been displayed on the screen, Windows stores it in a matrix of pixels, or tiny dots. It is impossible for the screen reader to interpret this information or to determine what is text and what is a picture. Windows-based screen readers intercept all information as it is being sent by Windows applications to the screen and store it in a memory construct known as the off-screen model (OSM). The screen reader then reads from the OSM rather than from the graphical image drawn on the screen itself. 2. Identifying and announcing the function of Windows constructs. Windows maintains the type, or class, of each element in an application, and most screen readers are capable of retrieving this information and delivering it to the user. In a typical Windows dialog box there may be a button that the user must select to proceed with a task. The Windows screen reader can identify the item as a button rather than simply reading the text and color of the item along with other text. 3. Identifying graphics. Many Windows features are not labeled with text, but are simply displayed as icons or pictures on the screen. Windows screen readers label these graphics so that they can be spoken in meaningful terms. A picture of a waste basket can be labeled "Delete," for example. 4. Serving as a mouse or pointing device. Some features of Windows applications are available only by clicking or dragging with a mouse. Screen readers have the ability to: move the mouse pointer in straight rows and columns or by meaningful units such as words or characters; find specified text and place the mouse pointer on it; and provide keystrokes that simulate the clicking of a mouse button. 5. Providing the information efficiently: in an order and with terminology meaningful to the user.

In other words, the screen reader must provide an alternative interface to the user that gives efficient access. A screen reader that reads the entire screen from top to bottom may eventually divulge the essential information, but it may take several minutes to do so. Likewise, it must be easy for the user to determine which of the items being spoken is the "current" item and which is additional, essential information. For example, if the speech program reads an entire dialog box, which of the controls is the focused item?

Program Profiles

JAWS for Windows (JFW) provides a set of basic speech commands enhanced by sophisticated, program-specific scripts. These scripts fine-tune JFW for particular Windows applications. Henter-Joyce provides well- written scripts for many popular applications, and JFW's performance excels in these applications. JFW simplifies the reading of the screen by presenting the screen as a series of lines of text, even when the text is scattered and not actually linear. This process gives the user a reliable way to read the current line in a word processor and also the focused item in a dialog box. This approach can be a problem, however, when nonlinear information is forced together--such as articles on the Web displayed in newspaper-style columns. JFW supports numerous refreshable braille displays and several languages. Eloquence software speech is included, making it possible for casual computer users with sound cards to avoid the expense of buying a hardware synthesizer. Window-Eyes was designed to work with a wide variety of applications out of the box. A large number of commands and options are available to fine-tune its function. These are accessed by going to the Window-Eyes menu and are easily changed by navigating through menus and dialogue boxes. Some basic settings--such as speech rate, pitch and volume--can be adjusted from the keyboard. Window-Eyes includes Microsoft's SAPI speech, making it possible for the casual computer user to avoid the expense of a hardware synthesizer. WinVision provides a large number of commands for navigating the Windows environment. Speech settings can be changed easily in a pop-up menu. Configuration files are provided for Word, Internet Explorer, Excel, and CD-Player. WinVision's command structure relies heavily on the control and alt keys. This causes many conflicts with Windows keyboard commands. Winvision's manual does not adequately explain a lot of its powerful features. outSPOKEN for Windows (OSW) provides the user with a small set of basic commands for navigating and reviewing text. Speech rate and a few other settings can be changed with hot keys. Other settings are changed in a pop-up menu. There are no program-specific configuration files, and no tools for creating them. As a result, all settings--such as verbosity or punctuation level--are global. Unlike most screen readers, OSW's Find command works in menus, making it a powerful tool for advanced users and less confusing for beginners. OSW supports braille displays and provides Grade 2 braille. Window Bridge offers speech output for both DOS and Windows in a single package. The advantage is that Window Bridge users do not need to buy a separate DOS screen reader if they want to run DOS applications within Windows. The disadvantage is that users already familiar with a particular DOS screen reader will not be able to use it. Window Bridge also supports braille displays. Its many prewritten configurations greatly improve its performance in popular applications. Tools and features are also provided that allow users to create and save custom configurations and to fine-tune those provided. Many Window Bridge features can be accessed with hot keys, which save time for experienced users but can contribute to key conflicts for beginners.

Problems

Problems with the OSM. Although the OSM makes the information on the screen available to the screen reader in a way that is otherwise not possible, it presents its own set of problems. The OSM can become outdated when the screen is updated and the OSM is not. It is also possible for information that is not actually on the screen to find its way into the OSM, meaning that the speech user hears information that the sighted user does not see. Incompatibility with Windows applications. Many Windows applications use conventional Windows programming techniques and consequently work at least relatively well with any good Windows screen reader. However, a large number of applications fail to adhere to these conventions and are consequently not completely compatible with screen readers. If a Windows application uses its own controls instead of those that are part of the Windows system--such as buttons, edit boxes, and check boxes--Windows screen readers have nothing meaningful to report to users when they encounter these "custom" controls. If these controls are not labeled or if the labels are painted characters instead of normal text, the screen reader cannot read them. Problems with stability and reliability. Partly because Windows screen readers interface extensively with the operating environment, bugs and problems in Windows, an application, or a screen reader may result in "crashes." Users of screen reading technology should expect to have more technical difficulties than many other users. lack of advanced efficiency features. At the beginning of Windows screen-reader history, these programs generally had generic, one-size-fits-all interfaces that read most information in most applications without much user-configurability. In recent years developers have refined their strategies to make their programs more efficient to use and have added many user options to allow for even greater efficiency at the user's work site. Still, no screen reader has all of the options necessary to make it an efficient tool for every user using every application. Some lack even basic configurability.

Testing of WINDOWS-BASED SCREEN READERS IN AFB'S PRODUCT EVALUATIONS LAB

Testing was performed between May 1999 and February 2000. Each program was tested on a Pentium 500 with 128 MB of memory, using a DECtalk PC synthesizer. Additional testing was done on slower machines with DECtalk Express, Audapter, and Double Talk PC synthesizers. Screen Readers Tested: JAWS for Windows, Henter-Joyce, Inc.< Window Bridge, Syntha-Voice computers, Inc.
Window-Eyes, GW Micro, Inc.
WinVision, Arctic Technologies International
outSPOKEN for Windows, Alva Access Group
HAL, Dolphin Computer Access

Tested Items

A rating was given for installation and documentation. Each program was then rated for performance with the word processing programs Microsoft Word 97 and 2000 and Corel WordPerfect 8.0 and with the Web browsers Microsoft Internet Explorer 5.0 and Netscape Navigator Gold 4.65. In order to evaluate a screen reader's configurability and the effectiveness of its advanced features, each was tested with aspects of Qualcomm's Eudora Pro, Microsoft Visual Basic professional, Microsoft Access 2000, Norton Utilities, Nullsoft Winamp, and other less common applications and utilities.

Advanced and efficiency features

Features that we especially looked for in evaluating a screen reader for advanced use included: a find feature that does not disturb the screen; the ability to automatically label graphics; a way to list links on a web page; flexibility in the order in which items are spoken; the ability to read system messages after an "illegal operation;" a generic way to read status lines in applications; the ability to read sentences and paragraphs; the ability to reclass problem controls; quick access to the system tray; and pixel by pixel drag-and-drop capability.


Go to previous article 
Go to next article 
Return to 2000 Table of Contents 
Return to Table of Proceedings


Reprinted with author(s) permission. Author(s) retain copyright.