ScanSoft TEXTBRIDGE PRO-MILLENNIUM BUSINESS EDITION User manual

COPYRIGHT
INFORMATION
Copyright © 1995–2000 by ScanSoft, Inc. All rights reserve . No
part of this publication may be transmitte , transcribe ,
repro uce , store in any retrieval system or translate into any
language or computer language in any form or by any means,
mechanical, electronic, magnetic, optical, chemical, manual, or
otherwise, without the prior written consent of ScanSoft, Inc., 9
Centennial Drive, Peabo y, Massachusetts 01960.
The software escribe in this book is furnishe un er license an
may be use or copie only in accor ance with the terms of such
license.
IMPORTANT
NOTICE
ScanSoft, Inc. provi es this publication “as is” without warranty of
any kin , either express or implie , inclu ing but not limite to the
implie warranties of merchantability or fitness for a particular
purpose. Some states or juris ictions o not allow isclaimer of ex-
press or implie warranties in certain transactions; therefore, this
statement may not apply to you. ScanSoft reserves the right to
revise this publication an to make changes from time to time in
the content hereof without obligation of ScanSoft to notify any
person of such revision or changes.
TRADEMARKS AND
CREDITS
TextBridge is a registere tra emark, an Smart Zones, Instant
Access OCR, an Custom Proof are tra emarks, of ScanSoft, Inc.
Excel an Word are tra emarks; Windows an FrontPage are
registere tra emarks of Microsoft Corp.
WordPerfect is a registere tra emark of Wor Perfect Corp.
Other terms use in this manual are the tra emarks of their
respective hol ers.
Animate character esigne by Dreamlight Incorporate .
www. reamlight.com.
Portions of this pro uct copyright © 1994–2000, Inso Corporation.
Authors: Lois West an Beth Pa ock
© SCANSOFT, INC.
9 Centennial Drive
Peabo y, Massachusetts 01960
TextBridge Pro Millennium Business Edition User’s Guide
Part Number 00–009733–00
April 2000

TextBridge Pro Millennium Business Edition User’s Guide iii
CONTENTS
PREFACE
About This User’s Guide ............................... vii
Organization of this user’s guide..................... viii
Documentation conventions ......................... ix
Related Documentation ................................ ix
Technical Support ..................................... x
1INTRODUCTION TO TEXTBRIDGE
asic OCR Concepts ................................. 1–1
Features and enefits ................................ 1–3
New Features ................................... 1–3
Enhanced Features............................... 1–4
Other Features .................................. 1–7
Documents Text ridge Can Recognize ................ 1–8
Input Image File Formats ............................. 1–9
Output Text File Formats ............................ 1–10
Output Image File Formats........................... 1–12
Where to Go From Here.............................. 1–13

iv TextBridge Pro Millennium Business Edition User’s Guide
2INSTALLING AND SETTING UP TEXTBRIDGE
What Comes with Text ridge .......................... 2–1
Supported Scanners.................................. 2–2
Installing and Testing Your Scanner ..................... 2–3
System Requirements ................................ 2–4
efore Installing Text ridge ........................... 2–4
Uninstalling a Previous Version of Text ridge .......... 2–4
Using Text ridge with Pagis ....................... 2–6
Learning about Text ridge before you install it ......... 2–6
Installing Text ridge ................................ 2–7
Scanner Setup ...................................... 2–8
Setting Up Instant Access to Text ridge .................. 2–9
Updating your Text ridge Software .................... 2–10
Uninstalling Text ridge Pro Millennium usiness Edition ... 2–10
Where to Go From Here.............................. 2–11
3 OCR AND BASIC TEXTBRIDGE OPERATIONS
What is Text ridge OCR? ............................. 3–1
Page types ..................................... 3–2
Page sources .................................... 3–3
Recomposition .................................. 3–4
Running Text ridge Standalone and Instant Access ......... 3–5
Standalone Program .............................. 3–5
Instant Access .................................. 3–6
Improving Page Recognition with Settings ................ 3–7
Page Type Settings ............................... 3–7
Text Document Settings .......................... 3–11
Recognizing Other Languages ......................... 3–13
Language Installation............................ 3–13
Language Processing ............................ 3–13
Where to Go From Here.............................. 3–15

Table of Contents v
4LEARNING TO USE TEXTBRIDGE
efore eginning to Process a Document.................. 4–1
Using Text ridge to Process a Document ................. 4–2
Starting Text ridge.................................. 4–3
Using Automatic Processing ........................... 4–4
Using Manual Processing ............................. 4–6
Performing asic Operations........................... 4–8
Selecting the Page Source.......................... 4–8
Selecting the Page Type ........................... 4–9
Previewing the Page ............................. 4–10
Zoning the Page ................................ 4–12
Proofreading the Document ....................... 4–14
Saving the Document ............................ 4–16
Getting Help While Using Text ridge ................... 4–17
Using the Welcome Window ....................... 4–18
Using the Show Me How Window ................... 4–18
Using Tips .................................... 4–20
Getting Information from Help ..................... 4–20
Using the Text ridge Web Site..................... 4–21
Where to Go From Here.............................. 4–22
5SAMPLE SESSIONS WITH TEXTBRIDGE
Using the Sample Documents .......................... 5–1
Session 1: Recognizing a Simple Document Using
Auto Processing .............................. 5–6
Session 2: Using Instant Access to Text ridge ............ 5–12
Session 3: Recognizing a Complex Document Using
Manual Processing........................... 5–17
Session 4: Processing Text, Pictures, and a Table .......... 5–25
Where to Go From Here.............................. 5–32

vi TextBridge Pro Millennium Business Edition User’s Guide
6ADVANCED SAMPLE SESSIONS
Session 1: Processing a Document to Use in a Database ...... 6–1
Session 2: Using Zone Templates and Page Types ........... 6–6
Session 3: Training Text ridge OCR .................... 6–11
Where to Go From Here.............................. 6–16
INDEX

TextBridge Pro Millennium Business Edition User’s Guide vii
PREFACE
ScanSoft, Inc. welcomes you to TextBridge Pro Millennium
Business Edition for Windows® 95, 98, 2000, and Windows
NT® 4.0.
T e documentation t at comes wit TextBridge provides you wit
t e information you need to operate TextBridge. T e
documentation includes t is user’s guide, a Help system, and
Release Notes. ScanSoft invites your comments about t e
information provided in t e documentation.
T e documentation is part of an extensive user assistance
program designed to provide you wit information you may need
to understand and use TextBridge. T e section “Getting Help
w ile using TextBridge” provides furt er information about user
assistance.
Before going on to find out more about TextBridge, please read
t is preface because it describes t ese important items:
◆ About t is user’s guide
◆ Related documentation
◆ Tec nical support
ABOUT THIS USER’S GUIDE
T is user’s guide is a reference tool t at provides information
about TextBridge. It is for users wit a wide range of computer
experience. It assumes t at you are familiar wit t e
management and operation of your computer and Windows.
T is manual is provided bot in print and electronic form. T e
entire user’s guide is provided as a digital document in Adobe®
Portable Document Format (PDF).

viii TextBridge Pro Millennium Business Edition User’s Guide
To view t e user’s guide in PDF format you need Adobe Acrobat
Reader, w ic is installed wit TextBridge, unless you already
ave it on your PC. You can access t e user’s guide from t e
installation menu and t e TextBridge Help menu, or you can open
it from Adobe Acrobat Reader. After you open it, you can view it
on your PC and print all or part of it using Adobe Acrobat Reader.
Organization of this user’s guide
T e user’s guide organization is as follows:
T e Table of Contents at t e beginning of t e user’s guide
describes t e basic information in t is book and elps you to find
general information quickly.
T is Preface describes t e documentation provided wit
TextBridge and tec nical support.
Chapter 1 “Introduction to TextBridge” discusses TextBridge
features. It also describes basic OCR concepts, documents
TextBridge can recognize, supported scanners, and input file
formats TextBridge can read and output file formats to w ic
TextBridge can save t e recognized text.
Chapter 2 “Installing TextBridge” describes w at comes wit
TextBridge; system requirements, installation, Instant Access set
up, and TextBridge uninstall.
Chapter “OCR and TextBridge” explains t e basic
TextBridge functions t at enable it to recognize and OCR your
documents.
Chapter 4 “Learning to Use TextBridge” describes t e basic
processes of using TextBridge.
Chapter 5 “Sample Sessions with TextBridge” walks you
t roug several practice sessions designed to elp you to learn
and use t e important features of TextBridge.
Chapter 6 “Advanced Sample Sessions” describes more
complex and less frequent uses of TextBridge.
T e Index provides a compre ensive list of topics to assist you in
quickly locating t e specific information you need.

Preface ix
Documentation conventions
TextBridge documentation uses certain grap ical elements and
formatting to emp asize information and give more meaning to
text.
Table 1: Documentation Conventions
bold Introduces a new term or t e first use of
an important term in a c apter. It is
sometimes used to denote strong
emp asis.
italic Denotes titles of ot er user’s guides or
books and generic representations of file
name entries in examples; for example,
filename
monospace Denotes text t at appears on t e
computer screen suc as examples, menu
text, and messages plus actual file
names.
“ ” (quotes) Denotes titles of c apters and sections in
t is user’s guide.
☞Introduces tips t at provide useful
information about a procedural step or
system function.
Note Introduces information of note about t e
current subject.
RELATED DOCUMENTATION
TextBridge provides a compre ensive set of printed and digital
documentation designed to assist you in learning and operating
t e product. T e documentation provided wit TextBridge covers
all aspects of installation and operation.
Note Information provided in individual documents is not duplicated in
ot er documents except for basic information about TextBridge. If
you do not find t e information you want in a particular
document, please c eck anot er. For example, if you do not find
information you want in t is user’s guide, look for it in t e Help
system

xTextBridge Pro Millennium Business Edition User’s Guide
Refer to t e documentation in t e following list for information:
◆ Online Release Notes. Before or after you install TextBridge,
read t e Release Notes. T ese provide t e most up-to-date
information about TextBridge. T ey describe tec nical
information, including specifics about using a particular scanner.
Release Notes also include information unavailable at t e time
t at t e user’s guide and Help were finalized. During installation
you can access t e Release Notes from t e installation menu.
After installation you can access t e Release Notes from t e
TextBridge Program menu in t e Start menu.
◆ Help. T e Help system provides you wit detailed information
about using TextBridge. It includes instructions on ow to get
started in TextBridge, step-by-step procedures for most operations
and user tips. Context-sensitive Help is always available by
pressing F1 from any menu command or dialog box.
◆ Online User’s Guide. An online version of t e complete user’s
guide is provided in Adobe Acrobat format (.pdf). You can access
t e user’s guide from t e installation menu and t e TextBridge
Help menu, or you can open it from Adobe Acrobat Reader.
◆ Printed User’s Guide. A printed version of t e user’s guide is
provided. T e user’s guide provides you wit basic information
about OCR and ow TextBridge performs OCR, information about
installing TextBridge, information about ow to run TextBridge
and improve its performance, and tutorials to elp you learn t e
basic-to-advanced use of TextBridge.
Note You may also need to refer to additional publications, suc as t e
manufacturer’s documentation for your scanner.
TECHNICAL SUPPORT
If you s ould experience problems wit TextBridge t at you
cannot resolve on your own using t e documentation and
software, contact TextBridge Tec nical Support at t e following
Web site:
www.scansoft.com

Preface xi
T e ScanSoft Web site provides a link to TextBridge pages,
including Tec nical Support wit Frequently Asked Questions,
tec nical information bulletins, and a problem report form.
Before sending a problem report form to ScanSoft Tec nical
Support, be sure to visit FastTrack, ScanSoft’s electronic support
system on t e web site. Using an Intelligent Expert Reasoning™
met odology, FastTrack delivers intuitive self-service via t e
Internet and allows you to successfully resolve your problems
online 24 ours a day, seven days a week. Software downloads,
product upgrades and product updates are also available in
FastTrack.
Additional information about contacting TextBridge Tec nical
Support is provided in t e TextBridge Help menu.
If you must contact ScanSoft Tec nical Support, t e following
information will elp in solving t e problem:
◆ Your software version number
(T is is on t e back of t e CD envelope and in t e Help menu
under About TextBridge.)
◆ Your software serial number
(T is is t e serial number on t e back of t e TextBridge CD-ROM
envelope and in t e Help menu under About TextBridge.)
◆ Your scanner make and model
◆ A description of t e steps t at led up to t e problem
◆ If TextBridge generated an error message, a verbatim description
of t e error message or its number and w en it appeared.

TextBridge Pro Millennium Business Edition User’s Guide 1–1
1
INTRODUCTION TO
TEXTBRIDGE
Welcome to ScanSoft’s TextBridge Pro Millennium, optical
character recognition (OCR) software for Microsoft Windows® 95,
9 , 2000 and Windows NT® 4.0.
This chapter provides an introduction to TextBridge including:
◆ Basic OCR concepts
◆ Features and benefits
◆ Characteristics of documents TextBridge can recognize
◆ Input image file formats
◆ Output text file formats
◆ Output image file formats
BASIC OCR CONCEPTS
OCR technology enables you to convert paper documents into
fully editable text with images on your computer. Originally, OCR
technology performed simple character recognition of text
characters, numbers, and symbols. Today, TextBridge OCR
includes full document recognition including recognizing text
plus formatting such as headlines, multiple columns, tables, and
running headers and footers and capturing photographs and line
drawings. TextBridge even retains the layout of the original
document as much as possible.
You can use TextBridge to scan and convert printed pages to
text documents for your word processor, spreadsheet program,
web browser, database program, or other text application. Pages
may be from most sources, including computer printers, fax
machines, photocopiers, magazines, and newspapers. Pages can be
black and white or color. TextBridge can also recognize standard
page image files from fax modems, image applications, and other
sources.

1–2 TextBridge Pro Millennium Business Edition User’s Guide
Using the latest document recognition technology from ScanSoft,
TextBridge OCR uses its recomposition capability to produce a
fully editable electronic document with the original pictures and
document layout (Figure 1–1).
Original
document
Recomposed
document in
word processor
Figure 1–1. TextBridge document recomposition
In most cases, TextBridge understands your original document’s
format and maintains the layout, including columns, headers,
footers, pictures, and picture captions. Pictures can be black and
white, grayscale, or color.

Introduction to TextBridge 1–3
Recomposition is possible only if your text program supports
pictures and layout. For example, recomposition is supported in
Microsoft Word and Corel WordPerfect but not in Notepad. Forms
and documents created in desktop publishing programs are
usually too complex for recomposition by TextBridge as well as
your word processor. As a result, the text and pictures are
retained but the full layout is not.
FEATURES AND BENEFITS
TextBridge offers many features designed to make it easy to use
and increase your productivity. Whether you need to capture a
simple one-page letter, a magazine article, a spreadsheet, or a
long transcript, TextBridge can save you valuable time and effort.
In addition, TextBridge provides all the capabilities that
experienced OCR users expect.
With TextBridge, you can import most paper documents or
document image files to your computer. TextBridge attains the
highest degree of OCR accuracy and provides the output in fully
editable form in your favorite program. Many of these features
and benefits are described in more detail in the user’s guide and
Help.
New Features
TextBridge Pro Millennium Business Edition offers these new
features to increase your productivity:
◆ Windows 2000 Certification. Makes use of latest Windows
technology to assure a consistent user experience and a more
reliable and manageable application.
◆ pdated scanner support. Includes latest Scanner Wizard hint
file for easy setup of popular scanners.
◆ Live updates. Stay up-to-date with the latest product changes
from the ScanSoft, Inc. web site.
◆ Instant Access™ to FrontPage® 2000 and The Print Shop®
ProPublisher 2000. Use TextBridge Instant Access to scan,
recognize, and paste text and pictures directly into your
FrontPage and Print Shop documents.

1–4 TextBridge Pro Millennium Business Edition User’s Guide
Enhanced Features
In addition to the new features, TextBridge offers enhanced
features that were available in previous versions. These features
were available before and are even better now. They are described
in the following list:
◆ Instant Access. Start TextBridge within most Windows text
programs such as Word or Excel. After recognizing and converting
the page, TextBridge then automatically pastes recognition data
(text and pictures) directly into the program’s open document.
◆ OCR accuracy. Dramatically save time and eliminate retyping.
◆ Color and grayscale pictures and text. Recognition and
output of color and grayscale pictures. Recognition of color text
and text on a color or shaded background and output of black on
white or white on black.
◆ Table recomposition. Advanced analytical capability results in
very accurate table reformatting. Ability to edit the entire table as
well as individual cells for improved recognition. Cell table
recomposition is supported even if you do not choose to retain
layout.
◆ Flexible multi-page document handling. Ability to view and
manipulate the pages of a document using the page thumbnails.
Zone multiple pages before recognition. Process the pages of a
document in any order. Delete, rearrange, and re-recognize
individual pages. You can also control the output.
◆ Extensive language recognition. Ability to recognize many
Eastern, Central, and Western European languages.
◆ Multiple language recognition. Ability to recognize multiple
languages on the same page if all languages belong to the same
language group.
◆ sability and user assistance. Enhanced ease of use including
a redesigned user interface and extensive user assistance. User
assistance includes a multimedia assistant, information screens,
context-sensitive tips, status area messages, Help system, and
printed and online documentation.
◆ TextBridge Assistant. An easy-to-use assistant guides you
through each step of the most common TextBridge activities, such
as how to scan a page and send it to Word, recognize an image file,
and recognize just part of a page.

Introduction to TextBridge 1–5
◆ Convenient batch processing. The ability to select multiple
files and process each file separately plus the ability to schedule
processing for a specific time in the future.
◆ Integration with e-mail programs. Input to popular programs
such as Lotus cc:Mail, Microsoft Outlook, and America Online
(AOL).
◆ Integration with the latest scanners. TextBridge works with
the most recent scanners. The Release Notes and the ScanSoft
Web site at www.scansoft.com provides the latest information
about supported scanners and getting your scanner to work with
TextBridge.
◆ HTML 4.0 output and WYSIWYG capability. Output files in
the latest version of HTML and preserve the original look using
cascading style sheets.
◆ Dual page scanning. Scan both pages of an open book at the
same time but handle them as two separate pages.
◆ Easy database importing. Use of standard delimited text file
output that allows you to import data into many databases.
◆ ToolTips and What’s This? Help. Instant context-sensitive
information about commands, dialog boxes, and buttons on the
interface.
◆ Document recomposition. TextBridge offers true document
recomposition to retain your original page layout. It reproduces
multiple columns, tables, and pictures and keeps them in the
same location as they are in your original document.
For example, when you specify output to the Microsoft Word™ or
Corel WordPerfect® format, TextBridge can retain the original
document layout in fully-editable form, even for pages containing
tables, line art, reverse video, drop caps, insets, and pictures.
When you edit the document, the original text flow is maintained.
When you specify output to the Microsoft Excel™ or Lotus 1-2-3
format, spreadsheets and cell tables retain their original layout as
cell tables, not tabbed columns. When you edit the table
information, the lines move to fit.

1– TextBridge Pro Millennium Business Edition User’s Guide
TextBridge supports formats for the programs that retain page
layout in the following list:
◆ Internet Explorer
◆ Netscape
◆ Word 6.0, 7.0, 97, and 2000
◆ Word Perfect 6.0, 6.1, 7.0, .0, and 9.0
◆ Any word processor that supports RTF
Retaining pictures is independent of retaining layout. Some text
programs retain pictures even though they do not retain layout.
◆ Page Types. TextBridge provides many predesigned Page Types
to make processing easier and more efficient. You do not have to
go through a complicated process of determining and specifying
settings for common types of pages. These Page Types
automatically provide appropriate settings for the type of page you
want to process. For example, there is a Letter page type and a
Magazine page type that automatically activate settings for
improved results for letters and pages from magazines.
◆ Automatic zoning. TextBridge automatically zones your page
into text, picture, and table zones. You do not need to zone the
page manually.
◆ Zone editing. You can edit the automatically recognized zones to
further refine the zoning. Use zone editing to increase the
accuracy and efficiency of page processing by reshaping zones,
specifying the language, and renumbering zones.
◆ Built-in Proofreader™. After document recognition, you can use
the built-in proofreader to view and accept or correct any words
that TextBridge suspects may not be recognized accurately. The
proofreader provides suggestions from which you may choose.
◆ Dynamic OCR training. You can train TextBridge’s OCR to
improve recognition accuracy as the job progresses. Use dynamic
training with difficult documents, such as faxes or multi-
generation photocopies. TextBridge enables you to interact with
the OCR process by viewing then accepting or correcting its
automatic recognition decisions. The software actually learns
special symbols and words.

Introduction to TextBridge 1–7
◆ Output files to the latest version of programs. These include
Microsoft Word 2000, Excel 2000, FrontPage 2000, WordPerfect
9.0, and Adobe FrameMaker 5.0.
◆ Custom dictionaries. To improve recognition accuracy further,
you can create specialized word lists (scientific terminology,
proper names, acronyms, and so on) within TextBridge or in
ASCII text files and load them into TextBridge. You can also use
your Microsoft Word or Office custom dictionary with TextBridge.
Other Features
In addition to the features listed in the previous sections
TextBridge provides these other features.
◆ Broad scanner support. TextBridge supports most popular
desktop scanners with TWAIN (technology without an important
name) device interface standard.
◆ Image processing. TextBridge accepts a wide range of images
from a variety of sources for processing. Specifically, the program
imports and recognizes online document images in BMP, PCX,
DCX, TIFF, and XIF formats that originate from fax modems and
other sources. For more information, see the “Input Image File
Formats” section in this chapter.
◆ Deferred processing. TextBridge enables you to scan all the
pages of a document to a TIFF or XIF file, then later open the
image file for document recognition. You can also save all the
pages to a multi-page image file or save each page as a separate
file.
◆ Output text file formats including HTML. TextBridge
supports a number of output text file formats, including word
processor, desktop publishing, spreadsheet, HTML, and database
formats. Now you can process your text for publication on the
Web.
◆ Preview of page images. TextBridge provides a set of tools for
previewing page images before processing them. You can
manually define areas of page images as zones to be processed and
capture only the text, tables, or pictures you want. You can also
edit the automatic zoning by adjusting the text, table, and picture
zones.

1–8 TextBridge Pro Millennium Business Edition User’s Guide
◆ Zone templates. After you create a set of zones, TextBridge lets
you save and reload zone templates for new jobs. In this way you
can consistently process or ignore specific areas on the same type
of pages and save time without rezoning each page.
◆ Re-usable training data. After you interactively train OCR, you
can save the training data in a file. You can reload this training
file for similar documents of the same page type. Using this
training file assures the highest recognition accuracy without your
having to repeat the training.
◆ Two-sided document processing. If your scanner has a sheet
feeder, you can scan the fronts (odd sides) of the pages first, then
flip the stack, and scan the reverse (even) sides. When scanning
and recognition are complete, TextBridge automatically collates
the text and keeps it in the original order.
Documents TextBridge Can Recognize
TextBridge includes a number of advances developed by ScanSoft,
Inc. and at the Xerox Palo Alto Research Center (PARC).
Consequently, TextBridge provides highly accurate OCR and
format retention on the widest range of documents. TextBridge
can recognize documents with the characteristics in the following
list:
◆ Documents printed on typewriters, phototypesetters, and impact,
ink-jet, dot-matrix, and laser printers
◆ Photocopied, degraded, or dirty documents
◆ Documents with single- or multiple-column layouts
◆ Spreadsheets and cell tables
◆ Paper documents with black and white, grayscale, and color
pictures including photos and line art
◆ Page image files with black and white, grayscale, and color
pictures
◆ Dual page documents, such as bound books
◆ Multi-page documents
◆ Single- or multiple-page images from fax modems and other
sources

Introduction to TextBridge 1–9
◆ Hard-copy faxes
◆ Documents with point sizes ranging from 5-point to 72-point type
in practically any typeface
◆ Documents composed in any of many Eastern, Central, or Western
European languages as well as one or more of the languages
within one of these groups in the same document
INPUT IMAGE FILE FORMATS
The source of page images for TextBridge can be your scanner or it
can be image files. TextBridge can recognize the following types of
image file formats:
Image File Format File Name Extension
Windows bitmap .bmp
PCX .pcx
Multi-page PCX used in some fax
programs
.dcx
Tag image file format (including
Alacrity TIFF)
.tif, .ala
Delrina WinFax fax image files .fxr, .fxd,
.fxm, .fxs
eXtended image file .xif
Image files can be black and white (binary), grayscale, or color.
TextBridge can process images in resolutions from 72 to 900 dots
per inch (dpi). Recognition results are generally better from
grayscale images than binary images. For the most accurate
results, we recommend scanning grayscale images at 200 dpi. For
better results on difficult documents, we recommend scanning
grayscale images at 300 dpi; however, this requires more
processing time.
Note Refer to the ScanSoft Web site at www.scansoft.com for the latest
list of supported input image file formats.