User Manual
MacSpeech Dictate™
License Agreement
Copyright [c] 1998-2008 MacSpeech, Inc. and its licensors.
All Rights Reserved.
5. U.S. Gove rnment Restricte d Rights
The SOFT WARE PROD UCT is provide d with Restri cted Right s. Use, duplication or disclosure by
or to the United States Government is subject to restrictions as set for th in subpar agraph (c)(1)(ii)
of the Rig hts in Technica l Data and Computer Sof tware clause at DFARS 252.2 27-7013 and/or
subparagraphs (c)(1) and (2) of the Commercial Computer Software - Restr icted Rights at 48 CFR
52.22 7-19, as applicable. Manufacturer is MacSpeech, Inc. in Salem, New Ha mpshire, U SA.
IMPORTANT - READ CAREFULLY : Before installing this software carefully read the following terms and condit ions. This S oftwa re End-User License A greement (“License Agreeme nt”)
is a bin ding agreeme nt between you (either an individual or a single entity), the e nd-user, and
MacSp eech, Inc. (“MacSpeec h”) r egarding use of the soft ware accompanying this Agreement,
which includes computer soft ware (“SOFT WARE”) and accompanying documentation and may include “online” or electronic documentation and a license key (together “SOF TWARE PRODUCT”).
By in stalling, copying or otherwise using the SOFTWARE PRODUCT, by clickin g on “yes”, you
agree to be bound by the terms of this L icense Agreement. If you do not agree with the terms of
this Lic ense Agre ement, cl ick on “no”, and t he instal lation pr ocess wi ll not cont inue.
1. Grant of License
If you are an individual, this Licens e Agreement grants you (“ Recipient”) a non -exclusive and
non-transfer able r ight to use one copy of the SOFT WARE PRODUCT in the manner d escrib ed
in thi s License Agre ement. If you a re an entity, this Lice nse Agreement g rants yo u (“Re cipient ”)
the right to designate o ne individual within Recipient’s o rganization to have the sol e ri ght o n a
non- exclusive and non -trans ferable basis to use one copy of the SOFTWARE PRODUCT in the
manner described in this Lic ense Agreement.
2. Use Limitations
(a) Single Mac hine
Recipi ent may use the license d copy of the SOFTWA RE on a single computer. The SOFT WARE
is in “u se” o n a computer when it is lo aded into tempora ry memory (i.e. RAM) or installed into
permanent memor y (e.g. hard dis k, CD- ROM or othe r storage device) of that computer. Use on a
“single computer” permits you to transfer the SOFT WARE from one computer to another comp uter
provided the SOFTWARE is i n use on only one computer at a time.
(b) Copies Permitte d
Recipi ent may make one copy of the SOFTWARE solely for backup or archival purposes provided
that Recipient reproduces all copyright , confidenti ality and other pro prietary notices that ar e on
the original copy of the SOFT WARE. Recipient may not copy the printed materials accom panying
the SOF TWARE.
(c) Transfer Prohibited
Recipi ent may not r ent, leas e, sell or otherwise trans fer the SO FTWAR E, whether on the media,
if any, or otherwise, nor any copies of the SOFT WARE, or any of the ac companying docum entation. Recipi ent may not disclo se, make availa ble, or otherwise redistribute all or any part of the
SOFT WARE PROD UCT or any cop ies there of to third parties. In addition, Recipient may not use
the SOFTWARE or any part th ereof, in any for m, in software or other product that i s or will be
distri buted to a thi rd part y.
(d) Limitat ions on Rever se Engineering/ M odific ations
Unless applic able law prohibits enforceme nt of this provision, Recipie nt shall not decompile,
disass emble, reverse engineer, or create derivative works of the SOFTWARE PRODUCT or
of any part of the SO FTWAR E PRODUCT. Re cipient may not alter or modify the SOFT WARE
PRODUC T or any part thereof.
3. Copyr ight Limitations
(a) Ownership
The SOFTWARE PRODUCT is licensed, not sold to Recipient, for use only under the term s of
this L icense Agreeme nt, and MacSp eech reserves all r ights not expressly gr anted to Recipi ent.
Recipi ent owns the media, if any, on whi ch the SOF TWARE PRODUCT is recorded. M acSpee ch
or its suppli ers owns the SOF TWARE and all copyright and other intellectual rights in the
SOFT WARE PRODUCT (including but not limi ted to images, “applets”, photographs , animations,
video, audio, music and text inc orporated into the SOFT WARE PRODUCT), the accompanying
printed materials , and any copies of the SO FTWAR E. Under the ter ms of this License Agreeme nt
Recipi ent receive s only a limited right to use the SO FTWAR E PRODUCT under all terms set forth
by Mac Speech, Inc., which are subject to change. The SOFTWARE PRO DUCT is prote cted by
United States copyri ght laws and internatio nal copyright treatie s, as we ll as by other intellectual
proper ty laws and treaties. Therefo re, Recipient mus t treat the SOF TWARE PRODUCT like any
other copyrighted materia l and not transfer or distribute the SOFT WARE PROD UCT to others.
(b) No Copie s
Except as expressly permitted under “Use Limitati ons” above, Rec ipient may not c opy the
SOFT WARE PRODUCT or acc ompanyin g writte n materials.
4. Term
This License Agreement is effec tive until ter minated. Recipient may terminate thi s License
Agreement at any time by destroying t he SOF TWARE PRODUCT and all copies the reof including any doc umentation. T his Li cense Agreement will ter minate immediately without notice from
MacSp eech if Recipi ent fails to c omply with any provision of th is Agreement. In such event,
Recipi ent must dest roy the SOFTWARE PRODUCT and all of its component pa rts, and delete,
remove, or oth erwise destroy all cop ies of the SOFT WARE PRODUCT t hat are in Recipie nt’s
possession or control including installed SO FTWAR E. All provisions of this License Agreement
relating to disclaimers of warranties, limitation of liability, remedies or damages, and all of
MacSp eech’s prop rietar y rights s hall sur vive termination.
6. Expo rt Restr ictions
The SOFTWAR E PRODUCT, inc luding technic al data, is subjec t to U.S. export control laws,
including the U.S. Ex port Administ ration Act and its as sociate d regul ations, and may be subject
to expor t or import regulatio ns in other countries. Recipient agrees to comply strictly with all such
regulations and ackn owledge s that Recipient has the responsibi lity to obtain licen ses to export ,
re-export or import t he SOFT WARE PRODUCT.
7. Limited Warra nty
With the exclusion of any other warranty, MacSpeech solely warrants for a period of six ty (60)
days from the date of delivery to Recipient (the “Warr anty Pe riod”) that any media on which the
SOFT WARE i s furnished, will b e free from defects in materials and workmanship under normal
use. Recipient ’s sole and exc lusive remedy and the entire liability of MacSpeech and its suppli ers hereunder will be, at Mac Speech’s option, rep air or replacement of the media, if repor ted (or,
upon request, returned) to M acSpee ch or its designee w ithin the Warranty Period. T his warranty
does not apply if such defective media is the result from accident, abuse, or misa pplicat ion. Any
replac ement will be warranted for the remainder of the original Warrant y Period or thirty (30) days,
whichever is longe r.
THE SOFTWARE AND ANY RELATED DOCUMENTATION IS PROV IDED “AS IS” WITHOUT
WARRANTY OF AN Y KIND. EXCEPT AS EX PRESSLY SET FORTH ABOVE, NO OTHER
WARRANTIES, EITH ER EXPRESS OR IMPLI ED CONDITIO NS, REPRESENTATIONS AND
WARRANTIES, ARE MADE WITH RESPECT TO THE SOFTWA RE PRODUCT INCLUDING
BUT N OT LIMI TED TO THE I MPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOS E, OR NON- INFRIN GEMENT, AND MACSPEECH AND ITS SUPPLIERS
EXPRESSLY DISCLAIM ALL OTHER WARR ANTIES NOT STATED HEREIN, EXCEPT TO THE
EXTENT THAT SUCH DIS CLAIMERS ARE HE LD TO BE LEGALLY INVALI D. RECIPIENT
ASSUM ES THE RISK AS TO THE QUALITY AND PERFORMANCE OF THE SOFTWARE. SOME
STATES D O NOT PERMIT THE EXCLUSION OF IMPLI ED WARR ANTIES, SO THE ABOVE
EXCLUSION MAY NOT APPLY TO YOU. THIS WARRANT Y GIVES YOU SPECIFIC LEGAL
RIGHTS A ND YOU M AY HAVE OTHER RIGHTS WHICH VARY FROM STATE TO STATE. IT
SHOULD BE UND ERSTOOD BY RECIPIEN T THAT SPEECH RECOGNITION IS INHERENTLY
A STATISTICA L PROCESS AND THAT RECOGNITION ERRORS ARE INHERENT IN THE
PROCESS OF SPEECH RECOGNITION.
8. Disc laimer
TO THE MA XIMUM E XTENT PERMITT ED BY APPLICA BLE LAW, IN NO EV ENT SHALL
MACSPEECH OR ITS SUPPLIERS OR THE OWNERS OF TH E RIG HTS IN THE SOFT WARE
PRODUC T BE LIABLE FOR ANY IN DIRECT, INCI DENTAL, CO NSEQUEN TIAL, SPECIAL,
PUNITI VE, OR OTHER DAMAGES WHATSOE VER, IN CONNECTION WITH OR ARISI NG OUT
OF THE USE OR INA BILIT Y TO USE TH E SOFTWARE P RODUCT (INCLUDING, W ITHOUT
LIMITATION, DAMAG ES FOR LOSS OF BUSINESS REVENUE, PROFITS, USE, DATA OR
OTHER ECONOMIC ADVANTAGE), HOWEVER IT ARISES, WHETHER FOR BREACH OR
IN TORT, EVEN IF MACSPEECH HAS BEEN ADVISED OF THE POSSIBILIT Y OF SUCH
DAMAGES. IN NO EVENT SHALL MACSPEECH’ OR ITS SUPPLIERS’ LIABILITY TO RECIPIENT
HEREUN DER, WHETHER IN CONTRACT, TORT (INCLUDING NEGLIGEN CE) OR OTHERWISE,
EXCEED TH E PRICE ACTUAL LY PAID BY RECIPIEN T FOR THE SOFT WARE PRODUCT. THE
FOREGOING LIMITATIONS S HALL APPLY EVEN IF THE ABOV E-STATED WARRANTY FAILS
OF ITS ESSENTIA L PURPO SE. BECAUSE SOME STATES AND JURISDICT IONS DO NOT
ALLOW TH E EXCLUSION OR LI MITATION OF LIABILIT Y, THE ABOVE LIMITATION MAY NOT
APPLY TO YOU.
9. Speci al Provisi ons
This License Agreem ent shall be governed by and interprete d under the laws of the Commonwealth
of Massachusett s, United St ates of Amer ica, wit hout regar d to conflicts of law provisions. By acceptin g this Lic ense Agreement, Recipient agrees to be subjec t to juris diction in Massachusett s,
and to commence any action arising under or relating to this Licen se Agreement in the appropriate
State or Fe deral Court in Mas sachusetts.
BY CONTINUING WITH THE INSTALLATION OF THE SOFT WARE RECI PIENT
ACKNOW LEDGES THAT RECIPIENT H AS READ AND UND ERSTOOD TH IS LICENS E
AGREEM ENT AND TH AT RECIPIENT AG REES TO BE BOU ND BY ITS TERMS. RECIPI ENT
FURTHER AGREES THAT THIS LICENS E AGREEME NT IS THE CO MPLETE A ND
EXCLUSIVE STATEMENT OF TH E AGREEMENT BETWEEN RECIPIENT AND M ACSPEECH ,
AND SUPERSEDES A NY PROPOSAL OR PRIOR AGREEM ENT, ORAL O R WRITTEN,
AND AN Y OTHER COM MUNIC ATIONS REL ATING TO THE SUBJECT MATTER OF THIS
Copyright (c) 1998-2008 Mac Speech, Inc. and it s licenso rs. All Rights Reserved. Mac Speech
SOFT WARE END -USER LICENSE AG REEMENT.
Dictate is a trademark of Mac Speech, Inc.
2 MacSpeech DictateLicense Agreement
Credits
MacSpeech Engineering:
Jeff Ganyard
Chris Hardy
Paul Herzog
Eric Hon-Anderson
Jim Kelley
Fernando Lins
Robert Stuller
Andrew Taylor
Colin Taylor
Chad Weider
Jonathan Wight
…and the rest of MacSpeech:
Sheila Ganyard
Stephane Gauthier
Jay Gonzales
T. Patrick Henebry
Carla Hernandez
Sara Jennings
Donald MacCormick
Dominique Maltais
Craig Nesbitt
Nathan Nesbitt
Anne Schwing
Michael Schwing
Brenda Shiepe
Jack Simonton
Carly Taylor
Special Thanks To:
Apple Inc. for the wonderful Mac OS X.
Nuance Communications, Inc. for their phenomenal speech engine.
Naomi Pearce and Ed Prasek for all their help introducing MacSpeech Dictate to the world.
And of course, thanks to everyone who has ever been involved with MacSpeech. MacSpeech Dictate certainly wouldn’t be
here without all your hard work and support!
Credit Given Where Credit Is Due:
Powered By Dragon®, the Dragon NaturallySpeaking® speech engine
from Nuance Communications, Inc.
The above are registered trademarks of Nuance Communications, Inc. and
are used here under license.
All rights reserved worldwide.
AquaticPrime Framework
Copyright © 2006-2008, Lucas Newman
All rights reserved.
BDAlias
Copyright © 2001-2008, bDistributed.com, Inc.
All rights reserved.
Copyright © 2007-2008 MacSpeech, Inc.
This user manual was partially written using MacSpeech Dictate.
For updates to this document, please see http://www.macspeech.com/support_center
CTGradient
Copyright © 2007-2008 Chad Weider.
Some rights reserved: <http://creativecommons.org/licenses/by/2.5/>
Log4Cocoa
Copyright © 2002-2008, Bob Frank
All rights reserved.
Shortcut Recorder
Copyright © 2006-2008, contributors to ShortcutRecorder.
Contributor details at <http://wafflesoftware.net/shortcut/contributors/>.
All rights reserved.
3 MacSpeech DictateCredits
Table of Contents
MacSpeech® Dictate License Agreement 2
Credits 3
Chapter 1: Introduction 5
Chapter 2: Installation 9
Chapter 3: Getting Started 10
Chapter 4: Training 14
Chapter 5: Dictation 20
Chapter 6: Controlling Your Mac 32
Chapter 7: Reference 38
Index 47
4 MacSpeech DictateTable of Contents
Chapter 1: Introduion
What Is MacSpeech Dictate?
MacSpeech Dictate is a new way to interact with your Mac.
Instead of typing your input, and clicking your mouse for
commands, now you can dictate your input and speak your
commands.
When you dictate words and phrases, what you say will appear on the screen, in virtually any Macintosh application.
The technology that enables this amazing feat translates
speech to text and characters almost anywhere you would
normally type.
MacSpeech Dictate can be used as a “third hand” to control
your Mac without using the mouse or keyboard. Speaking to
your computer is faster and can help you accomplish tasks
more easily than constantly moving your hands from keyboard to mouse.
MacSpeech Dictate is speaker dependent speech recognition
software. This means that in order to recognize your speech
so accurately, MacSpeech Dictate will be trained to recognize
your unique speech patterns.
Fortunately, MacSpeech Dictate is a quick learner. Initial
training can happen in a few short sessions, and the program
continues to learn as you go. Training data is stored in a profile
along with other things such as the type of microphone you
are using. MacSpeech Dictate can handle multiple profiles,
so others in your family or workgroup can use it, too (on the
same computer).
Will MacSpeech Dictate Replace My
Mouse And Keyboard?
Speech is a useful addition to the way you work with your
computer, but it isn’t a replacement. Some tasks will still be
efficiently performed using your mouse or keyboard. While
using MacSpeech Dictate, you will discover what combination
of speech, mouse, and keyboard use is appropriate for your
tasks.
Who Is MacSpeech, Inc.?
MacSpeech, Inc. people are passionate about the Macintosh
and experienced creators of speech recognition technology.
MacSpeech’s founder and senior engineers were key members of the teams that produced some of the first and some of
the best speech recognition products for Macintosh, including
Voice Navigator, PowerSecretary and iListen.
Being Mac-only means not having hands and creativity
tied by a requirement to maintain a common code-base
across computer platforms. As a uniquely Macintosh-based
company, MacSpeech, Inc. can take advantage of all the
Macintosh has to offer.
MacSpeech Dictate System
Who Should Use MacSpeech Dictate?
MacSpeech Dictate works well for many different people.
For a busy Mac user, MacSpeech Dictate can speed up
work when creating documents such as memos, reports, and
e-mails, or surfing the web and chatting.
Creative Professionals will love MacSpeech Dictate’s ability
to control the Mac interface, effectively allowing them to use
their voice as a “third hand.” MacSpeech Dictate can help
overcome mobility problems that make using a keyboard
difficult or impossible to use. Finally, those who have little or
no typing skills will benefit from MacSpeech Dictate’s ability
to free them from the “hunt and peck” method of typing.
No matter what type of user you are, you need to spend a
little time teaching MacSpeech Dictate the unique qualities
of your voice so the program can understand you better. As
you use MacSpeech Dictate it learns how your voice sounds,
your vocabulary and how you pronounce things. The more
you use the program, the better it will become at recognizing
your speech.
5 MacSpeech DictateChapter 1 — Introduction
Requirements
MacSpeech Dictate has the following minimum system
requirements:
• MacOSXversion10.5.2orgreater(“Leopard”)
or Mac OS X version 10.4.11 or greater (“Tiger”)
• 1GB(ormore)RAM.
• Intelprocessor-basedMac.
• 2GBoffreeharddrivespace.
• AMacSpeech-certiedUSBnoise-canceling
microphone.
Speech recognition takes a lot of horsepower. In order to
decipher what you are saying, your Mac needs to perform
thousands of calculations every second. Slow machines just
can’t keep up with the demand. So if you ever needed a
reason for buying that new top-of-the line computer with loads
of RAM, now you have one.
Microphones
MacSpeech Dictate supports microphones which connect to
the Macintosh through the USB port, either directly or through
a USB adapter.
Although today’s Macintosh computers appear to have a port
for sound input, this only works with devices that provide a
LINE IN signal, which is different from the signal produced by
a microphone. For this reason, you should use a high quality,
noise-canceling USB microphone with MacSpeech Dictate.
Tip
For best results, use a MacSpeech-certified
microphone. The iSight and the internal microphone
built into some Macintosh computers are not supported
because these do not have the necessary directional
and noise-canceling properties that are required to
obtain the best speech recognition accuracy.
may be small, but it packs a surprising amount of information,
as shown in Figure 1-1.
Figure 1-1
If you already have a noise-canceling microphone that is
not MacSpeech-certified, don’t panic; it may work. Dictate’s
Microphone Setup assistant can adjust the audio input level
of most microphones to work with your system.
If you use a microphone that is not MacSpeech-certified and
are getting accuracy below 95%, it is a good indication the
microphone you are using is not compatible with MacSpeech
Dictate.
If you received your microphone from MacSpeech along with
MacSpeech Dictate, it is certified to work with the software.
For information on microphones, please see the MacSpeech
Web site at:
http://www.macspeech.com/microphones
MacSpeech Dictate’s User Interface
By design, MacSpeech Dictate has a user interface that tries
to stay out of the way. The small Status window provides
control over the program’s main actions. The Recognition
window indicates what Dictate heard you say, and enables
the Phrase Training feature. The Available Commands
window shows which commands are available at a specific
point in time.
The MacSpeech Dictate Status Window
The Status window is MacSpeech Dictate’s main interface.
This window floats on top of all the other windows on your
screen, so it is always available to you. The Status window
The microphone button indicates your microphone’s status,
and controls whether it is on, off, or asleep:
• When the microphone is on, the button will
show a green circle behind a microphone. The
command Microphone Off or clicking the button will turn the microphone off. The command
Go To Sleep will make it asleep.
• When the microphone is off, the button will
show a red octagon in front of the microphone.
Click the button to turn the microphone on.
• Whenasleep,thebuttonwillshowapurpleoval.
The commands Wake Up or Turn Microphone
On, or clicking the button will turn the micro-
phone on.
Sound Input Level
The sound input level displays the strength of the audio input.
In general, this should be green with a little bit of black space
at the top when you are speaking.
Command Status Indicators
The command status indicators display the status of certain
commands. The top light is for capitalization commands (All
Caps/Caps/No Caps), the middle light for spacing commands
(NoSpace), and the bottom light for numeral commands
(Numeral/Roman Numeral).
MacSpeech Dictate Button
When pressed, the MacSpeech Dictate button will bring the
application to the front, just like clicking the MacSpeech
Dictate icon in the Mac OS X Dock.
6 MacSpeech DictateChapter 1 — Introduction
Speech Mode Indicator
The speech mode indicator text in the top right of the status
window tells you MacSpeech Dictate’s current mode. It will
say either “Dictation”, “Command”, “Spelling” or “Asleep.”
Each time you do this will activate the Phrase Training fea-
ture which helps Dictate continually learn more about your
speech.
Recognition Window Button
The recognition window button (picture of branching arrows)
is highlighted when there is a dictated phrase with multiple
recognition options available in the Recognition window.
Current Profile
The current profile text in the bottom right of the status window
is the name of the currently active user profile.
The Recognized Text Indicator
As you speak, the words as MacSpeech Dictate recognizes
them appear below the Status window in white, shadowed
text. Whether this text appears is determined by whether the
Show “Recognized Text” preference is checked in General
Preferences. The default is for the text to appear.
The Recognition Window
The Available Commands Window
The Available Commands window (Figure 1-3) provides a
list of commands that are available to be spoken. This list
varies depending on what is happening on the screen. You
will see different lists of available commands in this window
as the context changes.
Typing a word or phrase in the Search field in this window will
limit the display to only the commands containing that word
or phrase.
The Recognition window (Figure 1-2) shows Dictate’s
recognition of the words you dictated. This window will
appear dynamically during dictation depending on how the
Recognition Preferences are set. It includes a list of close
alternatives to choose from in case the program doesn’t
recognize your most recent word or phrase.
Figure 1-2
You can pick one of these options by saying “Choose <x>” or
“Pick <x>,” where <x> is the number of most accurate option.
Figure 1-3
User Modes
The specific things you can do using speech vary widely,
but they boil down to two main things: entering text and
controlling your Macintosh. In order to accomplish these
tasks, MacSpeech Dictate provides you with the following
operating modes:
Dictation Mode
You will probably use Dictation Mode the most. In Dictation
Mode, MacSpeech Dictate types what you say into a text
area in the active window. That text area can be in a word
processing document, a text field in a dialog box, or even text
that renames icons in the Finder.
All commands available in Command Mode are also available in Dictation Mode. In order to execute a command in
Dictation Mode pause briefly before and after saying the
command. For Example:
This is a sentence that has been dictated PERIOD
[pause] Select All [pause] Copy Selection
The above sequence will type the dictated sentence, then
select the whole document and copy it to the clipboard.
7 MacSpeech DictateChapter 1 — Introduction
Command Mode
Command Mode is used when you only want to control your
Mac by voice. In Command Mode, anything not recognized
as a command will be ignored. The advantage of using
Command Mode is that the commands will be more easily
recognized and will be less likely to be confused as something
you want dictated instead.
In Command Mode, you can click buttons, control windows,
launch applications, and use speech instead of pressing keys
on the keyboard or using the mouse. Many commands in
MacSpeech Dictate are written in AppleScript, the scripting
language built into Mac OS X. Commands can be global,
meaning they work everywhere, or application specific, which
means they only work within a particular application.
Spelling Mode
Spelling Mode allows you to spell a word that is not part
of the regular MacSpeech Dictate vocabulary. For example,
you might use this to correctly dictate names and other
proper nouns, proprietary terms, words in other languages,
or acronyms.
Figure 1-4
Sleep Mode
Sleep mode is a special mode where the microphone stays
on, but ignores everything except a command to resume
listening. You say Go to Sleep to put MacSpeech Dictate into
sleep mode, and Turn Microphone On or Wake Up to return
it to the last mode used before putting it to sleep.
Switching Between User Modes
There are several ways to switch between the user modes:
Speech
Say Command Mode, Dictation Mode, Spelling Mode or
Go to Sleep.
Menu
You can also switch modes by bringing MacSpeech Dictate
to the front and selecting the desired mode from MacSpeech
Dictate’s Speech menu.
Hot Keys
You can turn the microphone on or off and cycle between user
modes with customized keyboard shortcuts. To change your
Hot Key preferences, select Preferences from the Dictate
menu and click the Shortcuts icon in the toolbar (figure 1-4).
Any key can be used for a hot key, provided you use at least
one modifier (Command, Control, Option, etc.).
8 MacSpeech Dictate
Chapter 2: Inallation
Insert the CD titled “Application Disc”. A window like the one
in Figure 2-1 should appear. If it does not, double-click the
CD image on your desktop. Drag the MacSpeech Dictate
icon shown over to the picture of your Applications folder.
This will copy MacSpeech Dictate and its documentation to
the Applications folder.
Figure 2-1
Launch MacSpeech Dictate by opening your Applications
folder, then double-clicking on the MacSpeech Dictate icon.
Click the Register button to fill out the required information in
the Registration window (Figure 2-2). MacSpeech respects
your privacy, and will not share your information with anyone
else.
Data DVD
During the initial installation, set-up and training, MacSpeech
Dictate will ask for the Data DVD (in most versions titled
“MacSpeech English Data Disc”). This comes with the product, and is a separate disc from the Application CD. Please
have the DVD handy for when the program requests it.
Your Reiration Code
Registration Codes have 16 characters broken into
4 groups of 4 characters each by dashes. Here is an
example:
1A23-4B5C-678D-901E
This code should be on a sticker, on the sleeve for the
MacSpeech Dictate Program CD. After entering your
Registration Code you will receive a license file back
from MacSpeech’s servers.
Removing MacSpeech Dictate
To remove MacSpeech Dictate, follow these steps:
1) Open the Applications folder on your hard drive.
2) Delete the MacSpeech Dictate application, and close the
Applications folder.
Figure 2-2
Click the Register Now button to begin using the program.
(Registration requires an Internet connection.) A dialog box
will appear to confirm your registration has been accepted.
3) Open the Library folder inside your Home folder.
4) Open the folder called Application Support.
5) Look for a folder called MacSpeech; if it exists delete it.
6) Close the Application Support folder and open the
Preferences folder inside the Library folder.
7) Find the file called com.macspeech.dictate.plist and if it
exists, delete it.
8) You can delete profiles by searching for files on your harddrive with the extension .dictationprofile.
Troubleshooting Installation Problems
If you have problems with installation, please visit the
MacSpeech support web site at:
http://www.macspeech.com/support_center
9 MacSpeech DictateChapter 2 — Installation
Chapter 3:
Gettin Started
The first step is to get MacSpeech Dictate used to your voice.
You begin by setting up your microphone, and creating your
first Profile. MacSpeech Dictate requires you Enable access
for assistive devices in the Mac OS X Universal Access in
your System Preferences. If this option is not turned on you
will see the following panel:
Figure 3-1
Enable access for assistive devices is on, you may close
the Universal Access System Preferences window and click
the Continue button. MacSpeech Dictate will not allow you to
proceed until Enable access for assistive devices is on.
If the Universal Access setting is correct, you will see the
Connecting a Microphone to your Macintosh panel (Figure
3-3). If you have not registered yet, you will also see a panel
asking you to register or select your license key.
If you see this panel, click the icon labelled Show Universal
Access Settings. This will bring up the window shown in
Figure 3-2.
Figure 3-2
Make sure the check box at the bottom for Enable access for
assistive devices is checked. (Don’t turn on VoiceOver at
the top of the window by mistake unless you want your Mac to
talk back to you for every thing it does). Onc e yo u h ave verif i e d
Figure 3-3
The Connecting a Microphone panel gives you information
about connecting your microphone. Follow the directions for
connecting your microphone.
IMPORTANT
You must use a USB connection for your microphone.
It is fine to use a USB adapter. However, MacSpeech
Dictate does not support non-USB input connections,
such as a Mac’s internal microphone, direct analog
microphone jacks, built-in Bluetooth, FireWire, etc.
Positioning the microphone is very important. The microphone should not be directly in front of your mouth, because
direct noise from your breath sounds can affect accuracy.
For most people, the microphone should be one or two
fingertips from the corner of your mouth and off to the side
a bit.
You will be able to test and adjust the connection and positioning during the next steps of Getting Started. For now,
make sure the correct side of the microphone (sometimes
10 MacSpeech DictateChapter 3 — Getting Started
marked with a dot or the word “Talk”) is pointing towards your
mouth. When the microphone is connected and positioned,
click OK to close the Connecting a Microphone panel.
Creating a Profile
Now it is time to create a profile that will let MacSpeech
Dictate learn about your voice and your microphone. You will
see a blank Profiles Window (Figure 3-4).
Figure 3-4
TIP:
The “Microphone” pull-down menu may list something
like “AK5370,” “C-Media USB Headset” or “VXI 7.0.2.”
That’s OK. That is the firmware designation for the
USB adapter. (Your Mac has no way to otherwise
identify an analog microphone.)
Select the language model that best suits you from the
Language pull-down menu. MacSpeech Dictate supports
language models including US English, UK (British) English,
US Teen, Australian English, Indian English, and Southeast
Asian (Singaporean) English.
If you use a high-quality, noise-cancelling microphone such
as one from MacSpeech, select Standard Acoustics. If you
are using an Array microphone, click on the Advanced but ton
and then select “Array” from the pop-down menu next to the
word “Acoustics.”
When you have verified your choices, click the Create button.
After a short wait your profile is created. The first Microphone
Setup panel (Figure 3-6) appears.
To create a profile, click the “+” button. This will let you enter
a profile name, and select microphone type and language for
the profile (see Figure 3-5).
Figure 3-5
Select the type of microphone you are using from the
Microphone pull-down menu.
Figure 3-6
The Microphone Setup Volume Adjustment steps will ask
you to read a short bit of text out loud while MacSpeech
Dictate sets the audio levels for your microphone. Click the
microphone icon with the red octagon on it, to enable the microphone. This icon will turn into a green light, and the window
will display a box with text to read (Figure 3-7).
11 MacSpeech DictateChapter 3 — Getting Started
Figure 3-7
Read the text in this panel at a normal conversational volume
and pace. You should read until the text box disappears and
the panel changes. You will see the volume meter bar fill, and
the gain indicator slide as you read. If nothing happens after
reading the text, simply start reading the text again from the
beginning.
If the volume meter bar remains empty, or the text doesn’t
disappear after you read it three (3) times, you probably have
a sound input problem. See the Troubleshooting Sound
Input section at the end of this chapter for more information.
resonances from the bones in our head. Recordings lack
these extra resonances, so they sound different than we
expect. (If you don’t hear anything during the playback, check
to make sure your volume is not muted.)
Listen for static, excess noise, a hum, or any other odd sound.
If you hear any these of things, check your microphone’s
connections. If the connections seem OK, the microphone
may be faulty.
To hear the voice sample again, click the Play button.
MacSpeech Dictate is very good at determining the quality
of the sound input, but if you hear anything unusual, consider
re-sampling your voice. To re-sample your voice, click the left
arrow to go to the previous panel.
When you are satisfied with your voice sample, press the
Voice Training button to proceed to Voice Training.
TIP:
Microphone Setup can be used at any time, for example,
to adjust for room noise or microphone positioning.
Run this procedure by selecting Microphone Setup
from the Tools menu.
Checking the Manual Gain Setting b o x w i l l a l l o w y o u t o b y p a s s
automatic volume adjustment and set the gain manually.
The Finished recording panel (Figure 3-8) plays back some
of what was recorded so you can check the sound quality.
Figure 3-8
Chances are, you will think your voice sounds funny. That’s
normal; when we speak, we hear our voices with additional
12 MacSpeech DictateChapter 3 — Getting Started
Troubleshooting Sound Input
If MacSpeech Dictate doesn’t respond to your voice or doesn’t
seem to be working at all, check your sound input. The
problem can either be with the hardware (your microphone
or computer) or with software (the Macintosh sound input
settings).
Checking Your Microphone
Make sure the USB adapter is plugged into an open USB port
on your Macintosh. You may want to unplug it and plug it back
in to make sure, then try the following:
1). Quit MacSpeech Dictate if it is running.
2). Open your System Preferences and click on the
Sound icon.
3). Click on the Input tab and make sure your headset is highlighted in the input source list.
4). Click on the Show All icon in the upper left hand
corner of the System Preferences window, then
click the Speech icon.
5). Click the Speech Recognition Pane, then highlight the Listening sub-pane in the middle of the
window.
6). Select your headset from the pop-down menu
labeled “Microphone” towards the bottom of the
screen. Sometimes your headset may be indi-
catedbythermwaredesignatorofyourUSB
pod. This might say something like “AK5370.”
7). Click on the “Calibrate” button. In the next
window, say a few words and note if the signal
meter moves when you speak. If it does,
everything should be working correctly. Quit
System Preferences and re-launch MacSpeech
Dictate. Follow the on-screen instructions or the
instructions in the User’s Guide to set up your
microphone.
If you are not getting sound, you may have a defective microphone. Examine it to make sure there is no mute switch
set to off (some headsets include mute switches). Unplug the
microphone from your USB adapter and unplug the adapter
from your computer.
Re-plug everything and then restart your computer. If
possible, try the headset in a different Macintosh to rule out
any problems with your computer.
13 MacSpeech DictateChapter 3 — Getting Started
Chapter 4: Trainin
This chapter covers two methods of training MacSpeech
Dictate to recognize your voice, Voice Training and Vocabulary
Training. Along the way, we will share some insight into how
MacSpeech Dictate works, as well as tips for better recognition. We will also explain how to manage multiple profiles.
Voice Training and Vocabulary Training are not performed during normal use of MacSpeech Dictate for dictation. However,
a third method, Phrase Training, allows you to train the program through the Recognition window during everyday use.
Since it’s part of the standard dictation process, the Phrase
Training feature will be covered in Chapter 5 (“Dictation”).
Why Do I Need To Train MacSpeech
Dictate?
MacSpeech Dictate is speaker dependent, which means it
needs to be trained to get the best accuracy. You do this by
reading one or more training stories so it can adapt to your
unique vocal characteristics.
Luckily, training MacSpeech Dictate is as simple as reading
aloud and only takes a few minutes. The Voice Training part
of the program will guide you through the process.
tend to pepper your speech with lots of “ums,” “uhs,” “likes,”
and “you knows” MacSpeech Dictate will dutifully type what
it thinks you said.
Voice Training
Once you have set up your microphone and created a profile, MacSpeech Dictate automatically starts Voice Training
(Figure 4-1) so you can begin personalizing your profile. You
can re-enter training at any time by choosing Voice Training
from the Tools menu.
Speaking Properly To MacSpeech
Dictate
It is important to relax and speak in a normal conversational
tone. There’s no need to shout or project your voice. In fact,
doing so can be counterproductive and lead to poor accuracy.
It can also lead to vocal fatigue and voice strain. So relax!
It is also important to speak at a normal pace. MacSpeech
Dictate is a continuous speech recognition product, so it
relies on the context of words within phrases to obtain its
accuracy. If you insert… unnatural… pauses… between…
words, accuracy will actually decrease. (This is also why you
shouldn’t look at the screen while you are dictating. Waiting
for MacSpeech Dictate to catch up with you will result in these
unnatural pauses.)
Try to speak clearly. When you talk to people, they can fill in
the words you leave out, and they can compensate for words
you mispronounce or slur. Computers can’t do that.
One last thing: just as MacSpeech Dictate cannot insert words
you don’t say, it also cannot eliminate words or phrases you
say, but do not want in your documents. For example, if you
Figure 4-1
Click the microphone icon and read the text box on the panel.
The text you read turns green as it is recognized. Note that
you need to say punctuation, such as COMMA and PERIOD
(US only) or FULL STOP (all dialects). These commands are
spelled out and capitalized to help you here.
TIP:
After reading the first training story, MacSpeech
Dictate will become familiar with your reading voice,
which may be different from your dictation voice. For
an idea of how accurate MacSpeech Dictate is at this
point, try reading it something out of a magazine or
newspaper.
This will be mentioned a couple of times, but it bears repeating: You need to talk fairly slowly when first starting
Voice Training, but remember to read the text smoothly and
naturally.
14 MacSpeech DictateChapter 4 — Training
The text you read turns green as it is recognized
(Figure 4-2).
After a bit of processing (Figure 4-3), MacSpeech Dictate will
indicate when you are at the end of the training section.
Figure 4-2
MacSpeech Dictate automatically moves to the next panel as
you read. If you make a mistake, don’t worry; if MacSpeech
Dictate doesn’t understand you, the text will turn red
(Figure 4-2).
If the text turns red, pause for a moment, then continue reading starting at the text that is not green. If the program stops
on a particular word, try saying just the one word, or with only
a couple of words following it. If this doesn’t help, click the
Skip Word button to continue training.
TIP:
If you get a lot of red text, read fewer words, then
pause until the text turns green. Gradually increase
the number of words you read before pausing as long
as the text continues to turn green.
Figure 4-3
Adapting your voice samples is a time-consuming process. Be
patient; your computer just needs a little time. When adaptation
is complete, MacSpeech Dictate will let you know.
TIP:
It could take a few minutes for MacSpeech Dictate to
analyze your voice. Sometimes it can take about as
long as it took you to read the stories.
Just be patient. Do not Force Quit. Doing so may corrupt your profile and make it unusable.
As you complete each panel, the progress bar indicates
where you are in the story, and how far it is to the end. Keep
reading until you finish the “Welcome” story.
15 MacSpeech DictateChapter 4 — Training
Vocabulary Training
MacSpeech Dictate will learn more about your voice as you
use it for dictation. The process of Phrase Training through the
Recognition window (see Chapter 5) will help the program pick
up your speech patterns. But you also can have MacSpeech
Dictate concentrate on learning your communication style
and new words by analyzing existing text.
One way to do this is to use the Vocabulary Training menu
item to “feed” the program text documents. Another method
is to use the spoken command Train Vocabulary from
Selection for text highlighted in any application. You can also
cut and paste this text into Dictate’s Note Pad and choose the
Train Vocabulary from Selection... from the Tools menu.
TIP:
The Vocabulary Training methods improve accuracy
by helping MacSpeech Dictate learn about the way
you put words together. For this reason, it is preferred
that you use text that is written by you, not someone
else.
Figure 4-4
2. You should now see the Include/Exclude panel
(Figure 4-5). Determine which words here you
don’t want MacSpeech Dictate to learn. For
example, there might be proper names that are
not a usual part of your writing, abbreviations or
parts of URLs. For any word you do not want to
add, highlight it in the left column, then click the
“Exclude” button.
Here’s how to use these features to have MacSpeech Dictate
learn new vocabulary from text in documents:
1. You select the text containing new words you want
the program to learn. You can do this by one of
these methods:
Choose Vocabulary Training from the Tools
menu. Read the introduction screen, then click the
right arrow button. A screen will appear (Figure 4-4)
tospecifyles.ClicktheAdd... button, and choose
plaintext or Rich Text(.rtf) les to use. Clickthe
right arrow button.
OR
Within any application, highlight the text you want
Dictate to learn, then speak the command “Train
Vocabulary From Selection”.
OR
Paste text into Dictate’s Note Pad, then choose
Train Vocabulary From Selection from the Tools
menu.
Figure 4-5
3. To move a word you excluded back to the
Included list, highlight it in the right column and
then click the “Include” button. When you are
satisedwithyourchoices,clicktherightarrowto
haveMacSpeechDictateprocessyourlesinto
itsvocabulary(gure4-6).
16 MacSpeech DictateChapter 4 — Training
Loading...