Digital PDFs

EK-DTC03-OM-001

December 1985

222 pages

Original

92MB

view download

OCR Version

60MB

view download

Document:	DECtalk DTC03 Text-to-Speech System Owner's Manual
Order Number:	EK-DTC03-OM
Revision:	001
Pages:	222
Original Filename:

OCR Text

EK-DTC03-OM-001

DECtalk DTCO3
Text-to-Speech System
Owner's Manual

Prepared by Educational Services

Digital Equipment Corporation

1st Edition, December 1985

Printed in U.S.A.
The reproduction of this material, in part or whole, is strictly prohibited. For copy
information, contact the Educational Services Department, Digital Equipment Corpora-

tion, Maynard, Massachusetts 01754.

The information in this document is subject to change without notice. Digital Equipment
Corporation assumes no responsibility for any errors that may appear in this document.

This equipment generates, uses, and may emit radio frequency energy. The equipment
has been type tested and found to comply with the limits for a Class A computing device
pursuant to Subpart J of Part 15 of FCC Rules, which are designed to provide reasona-

ble protection against such radio frequency interference when operated in a commercial
environment. Operation of this equipment in a residential area may cause interference,
in which case the user, at his own expense, may be required to take measures to
correct the inteference.

B25A is a trademark of Western Electric.
BASIC is a trademark of Dartmouth College.
Touch-Tone and AT&T are trademarks of American Telephone and Telegraph

Company.
The following are trademarks of Digital Equipment Corporation, Maynard,
Massachusetts.

d[ilgli[t]a]1]

DECtalk

DEC

DECUS

DECmailer
DECmate

RSTS

DECwriter

RSX

DIBOL

UNIBUS

DECnet

MASSBUS

VAX

Rainbow

DECservice

PDP

VMS

DECsystem-10

P/OS

DECSYSTEM-20

Professional

Work Processor

U.S. FCC REQUIREMENTS

The U.S. Federal Communications Commission (FCC) classifies the DECtalk DTCO03 as
terminal equipment. FCC regulations require that you provide the business office of
your local telephone company with the following information before you connect
DECtalk to the telephone network.

e The particular lines(s) to which terminal equipment will

be connected (by tele-

phone number)

e The make, model number, and FCC registration number (See

the label on the

back of the equipment.)

e The ringer equivalence for the registered terminal equipment (See the label on
the back of the equipment.)

e The type of service needed (if not already installed)

The following list contains that information you need to give the telephone company.

Make: DECtalk

Model: DTCO03-AA

FCC registration: A0994Q-71152-AN-E

Ringerr equivalence: 0.2B
Type of service: 8 public switched lines in a USOC RJ21X connection arrangement
(See the DECtalk DTCO03 Text-to-Speech System Installation Guide.)

You must also notify the telephone company when you permanently disconnect terminal
equipment from telephone line(s).

You may not connect terminal equipment to a party line or coin-operated telephone
equipment.

If terminal equipment damages the telephone network, the telephone company can,

after notifying the customer, temporarily discontinue service. However, when prior
notice is not practical, the telephone company can temporarily discontinue service
immediately. In such cases, the telephone company shall
e promptly notify customers that service has been

discontinued

e give customers the opportunity to correct the situation

e inform customers of their right to bring a complaint to the FCC according to
Subpart E of Part 68 of FCC Telephone Equipment Rules.

CANADIAN DOC REQUIREMENTS

The Canadian Department of Communications (DOC) label on the DECtalk DTCO03
identifies certified equipment. This certification means that the equipment meets
certain
telecommunications network protective, operational, and safety requirements. The DOC
does not guarantee the equipment will operate to the user's satisfaction.

DOC regulations require that you provide the business office of your local telecommunications company with the following information before you install DECtalk.
Make: DECtalk
Model: DTC03-AA

DOC certification: 1921422A

Load number: 5B
Type of service: 8 public switched lines in a CA21A connection arrangement (See
the DECtalk DTCO3 Text-to-Speech System Installation Guide.)

Before you install this equipment, make sure it is permissible to connect it to the local
telecommunications company’s facilities. You must also install the equipment by using
an approved connection method. Be aware that complying with the above conditions

may not prevent degradation of service in some situations. Telecommunications company requirements do not allow you to connect their equipment to customer-provided

jacks, except where specified by individual telecommunications company tariffs.

Only authorized Canadian maintenance facilities, designated by the supplier, should

repair certified equipment. If you repair or alter certified equipment yourself,
or if the
equipment malfunctions, the telecommunications company has cause to ask you to

disconnect the equipment.

You should ensure (for your own protection) that the electrical ground connections for
the power utility, telephone lines, and internal metallic water-pipe system, if present, are
connected together. This precaution may be particularly important in rural areas.

CAUTION: Do not try to make such connections yourself. Contact the appropriate
electric inspection authority or electrician.

CONTENTS

Al SR e
INTRODUCTION .......... TP
PART1

GENERALDESCRIPTION .............. R

Xiii
1

1.2.2.2
1.3

Pushbuttons ..., e
The DECtalk DiagnosticTests .................... s

1.4

OperatingMode .............. e e

2
2
3
4
4
6
7
8
8
8
9

en e
.
i
HOWDECtalk WORKS .....
2
.....
...
..
......
TexttoSpeech
Converting
2.1
DECtalk Software Program ...,
2.2

10
10
11

HOW DECtalk PROGRAMMING WORKS .................

COMMUNICATINGWITHDECtalk ........... ... . ot
3
.....
... ...
Programming Guidelines ........
~
3.1
i
TypesofData...........coiiiiiiiiiiiiiii
3.1.1
Using Escape Sequences and ControlCharacters ......
3.2
Escape Sequences .......... e e ceve....
3.2.1
Escape Sequence Format .............. e e
3.2.2
.....
.. ..
Special Characters .....
3.2.3
...............
Character
(BS)
Effect of the Backspace
|
3.2.4

16
16
18
18
18
20
22
23

GETTING TO KNOW THE SYSTEM .............. [,
1
The DECtalk DTCO3 System ........ .o,
1.1
The DECtalk Module ...t
1.2
Module Configuration ................. e ...
1.2.1
iie e
i
Module Controls ...t
1.2.2
e
e
LEDS ...,
1.2.2.1

1.3.1
1.3.2

PART2

Power-Up Self-Test ........ ...t e
Routine Self-Test ...

CONTENTS

3.3

DECtalk-Computer Communication ......... e e

3.3.1

3.3.2

DECtalk Setups........ e e e e
Program Control ......
... .......
... .. .. ..... e

3.3.3

Data Synchronization .....
... ..
.....
... ... ...
. ... .

CHARACTERENCODING AND SETUP ESCAPE

SEQUENCES ...t
4.1

Selecting ASCI|I
Character Sets.......... .

4.2

Coding Standards ........................ e

4.3

CodeTables

.......... i

4.3.1

7-Bit ASClICode Table

4.3.2

8-BitCodeTable .......................... e

4.4

...................... e

CharacterSets ..., R

4.4.1

Designating Alternate Graphic Sets .......... e

4.4.2

Invoking Graphic Sets into GL and GR ... .. e

4.4.3

4.5

Character Processing ............... e

Control Functions ...............co

4.5.1

Escape Sequences .............c.o ..

4.5.2

Control Sequences ................. ee

4.5.3

Device Control Strings ...........coouu
e ..

4.6

Working with 7-Bit and 8-Bit Environments .............

4.6.1

7-Bitand 8-BitC1 Controls

4.6.2

Mode Selection (DT_MODE)

.......... e

MAINTENANCE AND TESTCOMMANDS

...,

.........................

5.1

Conformance Level Request (DECSCL) ................

Device Attribute Requests ......................e

5.2.1

Device Attribute Request (DA anary) ...............

5.2.2

Device Attribute Request (DA Secondary) ...........

5.3

Device Test and Status ............ e e

5.3.1

DECtalk Power-Up Status ................. ...
...

5.3.2

Device Self-Tests (DECTST) ...

5.3.3

Device Status Request (DSR) (Brief Report) .........

534

Device Status Request (DSR) (Extended Report) ....

5.3.5

Reset to Initial State (RIS)

5.3.6

Soft Terminal Rest (DECSTR) ................. e

53.7

NVR Control (DECNVR) ...,

......
... ... ...
...
.....

CONTENTS

6
TELEPHONE COMMUNICATIONS .................. e,
6.1
~
Telephone Management (DT_PHONE) ................

57
58

6.1.1

PH_ANSWER ........ R

N Py

6.1.2

PH_HANGUP ...

e ...

6.1.3
6.1.4
6.1.5
6.1.6
6.1.7

PH_KEYPAD ............ et
en e s
e .
PH_NOKEYPAD . .
i
e e ceeee e
PH_STOPPAD ..... ..t e.
PH _TIMEOUT ...
et et e
e e
e
PH_TONE__DIAL and PH_PULSE_DIAL ..........

62
62
B2
63
63

6.1.8

PH _WINK

i i
ittt e et e eee e
PH_ _NOWINK .
i e et e et e i e
Examples of Telephone Communication Sessuons ...
Keypad Mask (DT_MASK) .................... e

DECtalk PHONEMICSANDVOICES .....................

DECtalk PHONEMICINPUT ............ e e e ce...

6.1.9
6.1.10
6.2
PART3
7

65
B65
70

7.1

Phonemic Transcription ...

7.2

Pronunciation Errors ........
..
i, e

7.3

Introduction to Phonemic Theory ......................

7.4

Phonemes ...........ciiiiiieann. e, ..

7.5

Vowel and Consonant Phonemes ......................

Phonemic Correction the EasyWay ............ e
Automatic Phonemic Substitution for a

7.5.2

Mispronounced Word ..... R

7.5.3
7.54
7.5.4.1
7.5.4.2
7.5.4.3
7.5.5
7.5.6
7.5.6.1
7.5.6.2
7.5.6.3
7.56.4
7.5.6.5
7.5.6.6
7.5.7

VOWEIS .
e
e
e
Vowel Allophones . ..............ciiiiiiieiaee...
Allophones for R-Colored Vowels ....... e .....
Schwa Allophone . ......... ..., ...
Syllabic Consonants ................ e e
ConsonNants ........ciiiiii
e e ..
Consonant Allophones ............. ... .. e
DentalFlap .............. P e
Glottal-T ... e ......
Postvocalic-R ...
Postvocalic-L ...
e
GlottalStop ................. A
SN
Controlling Allophone Selection ..................
Silence Phoneme [ ] ....... e
T e

83
83
84
84
84
85
85
85
85
85
86
86
86
86

7.5.1

vii

CONTENTS

7.6

Stress and SyntacticSymbols .............
. ... ... ...

7.6.1

Primary Stress["] .............. P

7.6.2

Secondary Stress[

] ...oovvirnininnn... e, ...

7.6.3

Emphatic Stress [”]

..o

7.6.4

Unstressed Syllables ............c.o

7.6.5

Pitch Control [/, [NLI/N] cevevneee

...

7.6.6

Word Boundary .......ooiiiiiiiiiii
i iieie i

7.6.7

Syllable Boundary [-] .............. et

e et

7.6.8

Morpheme Boundary [*] ...,

7.6.9

Compound NouN [#
] .o

7.6.10

Beginning of Verb Phrase [)]

..........cc
o ...

7.6.11

Clause Boundary [,] ..o

7.6.12

Period[.]............. etaereaiee
e ta e .

7.6.13

Question Mark [ 2] ..o iiin

e e

7.6.14

Exclamation Point [!] ............. Cevedemen e

New Paragraph [+ ] ... i

7.6.15

7.7

Direct Control of Durationand Pitch ...................

7.7.1

Duration and Pitch [<>]........ Ce e ereressar e

7.7.2

SINGING .o e 100

7.8
8

Selected Readings .........e

e e 102

VOICE COMMANDS AND THE USER DICTIONARY ..... ee..... 1083

8.1

Speech Control ....... e

8.2

SpeechTimeout..................... et 104

ee e e 103

8.3

English Text........ e e eveEesemeiiai et 104

8.4

Speak Phonemic Text (DT_PHOTEXT) ................ 105

8.5

Stop Speaking (DT_STOP) ..o

8.6

Data Synchronization (DT_SYNC) ................. .... 106

8.7

Enable or Disable Speaking (DT_SPEAK) ............. 106

8.8

Indexing

.................... e

viiii e 105

eretrere e oo 107

8.8.1

Index Text (DT__INDEX) ............. e ieerirnenan 107

8.8.2

Index Reply (DT_INDEX_REPLY) ................ .. 108

8.8.3

Index Query (DT__INDEX_QUERY) ................. 108

8.9

Load Dictionary (DT_DICT)

8.10

Input Logging (DT_LOG) ..., 111

............ e

110

MODIFYINGTHEVOICES ... 112

9.1

Voice Characteristics . .........coiiiiiii it 112

9.2

SpeakingRate[ira__]

9.3

Pause Durations[;pp__Jand|[cp_] .......... e 116

viii

... 115

CONTENTS

‘Selecting a Standard Voice[ln_] ................ooal
Designing a Voice with Speaker-Definition

117

Changing Sex and Head Size .......................

122

9.5.1.3
9.5.2

Higher Formants, f4, {5, b4, and b5 e
Changing Voice Quality .............coooiiiiian.t.

125

9.5.2.1
9.5.2.2
9.56.2.3
9.5.2.4
9.5.2.5
9.5.2.6

Nopen Fixed, nf ....... ..o,
Laryngealization,la ......... ...t

9.4
9.5

_] ...e
Parameters[idv

9.5.1
9.5.1.1
9.5.1.2

9.5.3
9.5.3.1

9.56.3.2
9.5.3.3
9.5.3.4
9.5.3.5

9.5.3.6
9.54

9.54.1

9.54.2
9.5.4.3
9.5.5
9.5.6

122
s R
SX viiiiniin
SEX,
i i 123
Head Size, hS ...ttt

Breathiness, br ... 125
eeaa e 125
Lax Breathiness, Ix ......i iieae s
Smoothness,sm ........... e e
RiChNess, M ....c.oiiiiiiiiiiiiiiiiaiennans TPy

Changing the Pitch and Intonation of the Voice.......
Baseline Fall, bf .....ead e i e Ceimameenraneenn

ise,, v e e |
hr ...
HatR
Stress Rise, sr.........ocooiiiennn. P e
... e
Assertiveness,as................
e iieeea e e
ie ittt
QUICKNESS, QU ..o

Average Pitch, ap, and Pitch F’iange pr ..........

Changing Relative Gains and Avoiding Overloads ...

Loudness, 1o ...c.vviiiiriii i AU

Sound Source Gains, gv, gh, and gf .............. ..
Cascade Vocal Tract Gains, g1,902,9g3,and g4 ....
The [save] Parameter and [:nv] Voice ................
Summary on Speaker-Definition Parameters ........

9.6

Voice Command Syntax ...........ooiiiiiiiiiiiiiia.

HOWTOGETTHE MOSTOUTOFDECtalk ....................

136

Names, Part Numbers, and Alphanumemc Text ......

138

10.1

10.1.1
10.1.2

10.1.3
10.1.4
10.1.5
10.2
10.3
10.4

Developing Your Application .... R P 136

Direct Numeric Encoding ........... S e 138

Two-Character EnCoding . ..o ovvvnnennneanannnnns 138

Ending CommandsandData ........................ 139
Application Development Tips .................. ‘i 139
Developing an Advanced Application .................. 140
Optimizing the Quality of Spoken Text ................. 142

COmmMONEIors ......cooviiiiiiinenennn. eeaeedeneane 144

CONTENTS

APPENDIXA

DECtalk ESCAPE SEQUENCES, STATUS REPLIES,

AND PARAMETERS..., e 146
APPENDIX B

B.1
B.1.1

CONFIGURING THE DECtalk MODULE ............ 154
Setting the Configuration Switches .................... 154
Line Speed ....... ... . 156

B.1.2

Data Format ....... e e e 156

B.1.3

Keypad Mask ......................... e e 156
Busyout ......
.. ... ...
....
... .. e 157

B.1.4
B.2
APPENDIXC

C.1
C.2
C.2.1

c.2.2
C.2.3
C24
C.2.5

C.2.6
C.2.7

C3
C.3.1

C.3.2

C4
C.4.1

Testing the Module ........ e

e, . 157

ABBREVIATIONS ANDACRONYMS ................ 159

Text ProcessingRules ...
... 159
Number Processing ........... e .. 161
Part Numbers..................... R e 161
Cardinal Numbers ............. e 163
Ordinal Numbers ...... ..., 166
Fractions .........................e 166
MoONeY ..o 166
Dates ... e 167

TimeofDay............ e
e e e e, 167
Abbreviations ............ ..., e 167
Abbreviations Processed by DECtalk ..... e e 167

Abbreviations in the Built-In Dictionary .............. 168
Word Spellout Strategies ...............co
L. 170

C.4.2

Word Spellout Tests ... . 170
Multinational Characters ...............
...
....... 172

APPENDIXD

WORDS WITH TWO PRONUNCIATIONS ........... 173

APPENDIXE

DECtalk/DICTIONARY PHONEMIC

CORRESPONDENCE ..., 177
E.1

E.2

Webster's Ninth New Collegiate Dmctlonary ...... ....... 180
American Heritage Dictionary ......................... 181

APPENDIXF

TEXTTUNINGEXAMPLE ................0cuuii
. 182

APPENDIXG

BASIC PROGRAMEXAMPLE ....................... 185

CONTENTS

SERVICE AND DOCUMENTATION ................. 191
APPENDIXH
Digital's Services ........... RAP Ceeeheneaea 191
H.1
DECservice ............. S 191
H.1.1
Basic Service ........ VN S e e Y TS 192
H.1.2
192
A
Per Call Service ............. A
H.1.3
192
e
e
e
Carry-In Service .............i
H.1.4
DECmailer Service . ......oiiiieienieiiininennnns ... 193
H.1.5
Digital Field Service Information Numbers .......... 193
H.1.6
Related Documents ......... e e 194
H.2
H.3

Ordering Information .......... ..o

ee e
e et
D E K .o
IN

i, 195

e 196

FIGURES

1-1
1-2
1-3
1-4
2-1
3-1
3-2
3-3
3-4

3-5
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
4-9
6-1
6-2
6-3
6-4

DECtalk DTCO3 System Configuration .................
DECtalk DTC0O3 Communications Cabling .............
Accessing the DECtalk Modules ........... e e .
Module LEDs and Pushbuttons ..................... ...
Text-to-Speech Conversion ...,
DECtalk-to-Host Communications .......... e
Typical Escape Sequence Format .....................
Escape Sequence Representations .............. e
DECtalk and Host Computer Programming

2
3
4
b
11
17
19
20

Mapping 7-Bitand 8-Bit Tables ........................
7-Bit ASCliCode Table ..................... e e
Data Word Representation .......... e,
8-BitCodeTable ............iiiiiiiiiiiiiiiiiinen..
ASCIl_G CharacterSet ..................... e
DEC Supplemental CharacterSet .....................
U.K. National CharacterSet ...........................
L.
.......
Loading 8-BitCharacters .............o
..ot
...
...
........
Loading 7-BitCharacters
Phone Call Session with User Disconnect .............
Successful Dialout Session................... e
Busyout Session .......... ..., R e
NO ANSWEr SESSION . ..ottt ittt ittt e i

30
32
33
34
36
37
38
40
41
66
67
68
69

t e ...
e
INEEraCtiON .ott
Synchronizing DECtalk and Host Communications......

25
28

CONTENTS

7-1

Phonemes Classified by Their Position in the
Mouth Cavity ..... et

8-1

9-1
B-1
B-2

ma e s ee
e enean s ataae
Using the DT_SYNC and DT__ IMDELQUERY to

Coordinate DECtalk Communications ................. 109
The DECtalk Voices ...
i, e 113
DIP Switchpack Settings .................... e .. 155
DECtalk Module Controls ...........c
i co
i, 158

TABLES
3-1

Special Characters and Host Communications .........

4-1

Selecting 7-Bitor 8-BitMode ............... e

4-2

Invoking G-Sets ...........

4-3

Designating CharacterSets ...........................

4-4

DT_MODE Parameters ....... e
e
e,

5-1
5-2

Restoring DECtalk Default Operatmg Features ........
DECtalk Actions Performed at Resets .................

5-3

Self-Test Parameters ................ e

5-4

6-1

DSR (Extended Report) Error Codes e ....
DT_PHONE Parameters .............c
oo
oo
i,

6-2

Phone Status ReplyCodes ...............ccouuo
i,

6-3

DT_MASK Parameters .................coii..

7-1

English Vowel Phonemes ..............................

7-2

English Consonant Phonemes and Mlophones

7-3

Stress and Syntactic Symbols .................. e

7-4

9-1
9-2

9-3
9-4
A-1
A-2
A-3

A-4

C-1

C-2
D-1

E-1

Xii

........

87
ToneValues.........coooiiuuin
., e 101
Voice Modification Commands .............. e 114
New Voice Commands ............covuviuiinnn.
. 117
Speaker Definition [:dv _] Parameter Range ........... 119
Speaker Definition for All DECtalk Voices .............. 121

Escape Commands.................. e 147
DECtalk Status Replies.............cc.ooui
... e 150
Conformance Level and C1 Format Setting
Parameters .......... e e e
e e e 153
Dialing Characters .......................... e 153
Numeric Abbreviations Recogmzed by DECtalk ........ 165
Abbreviations Recognized by DECtalk ................. 168
Words with Alternate Pronunciations ................
.. 174
DECtalk and Dictionary Phonemic Alphabets .......... 178

INTRODUCTION

ABOUT THE DECtalk DTC03

Video terminals display information from a computer on a screen. Printers
display the same information on paper. These devices allow you to communicate with computers through the sense of sight.
The DECtalk DTCO03 is another device that allows you to communicate with

computers. However, this device speaks information in English. It allows you to
communicate through the sense of hearing.

The DECtalk DTCO3 is a text~to-—-speeeh system It converts computer text to

computer speech. This system can provide any mmputw with a human sounding voice.

ABOUT THIS MANUAL

This manual provides information about the DECtatk DTCO3 system for general

users and programmers. To make reading easier, this manual uses the term
DECtalk when describing system performance.
The manual is divided into three parts. Part 1 provides a general description of

the DECtalk DTCO03 and how it works. Part 2 describes how to use and program the DECtalk DTCO03 to work with a host computer. Part 2 also explains

the escape sequences you can use with the DECtalk DTCO03. Part 3 explains
the DECtalk DTC03 phonemics and voices.

The appendices at the back of the manual provide additional information to

help you use and program the DECtalk DTCO3.

Xili

INTRODUCTION

Part 1

General Description

e Chapter 1, “Getting to Know the System,”” describes the DECtalk DTC03
system and its modules. This chapter also provides general operating and

testing information.

e Chapter 2, “How DECtalk Works,” describes the DECtalk speech generating system and gives an overview of how the system operates.

Part 2

How DECtalk Programming Works

e Chapter 3, ““Communicating with DECtalk,”” describes how the DECtalk

DTCO03 communicates with a host computer through a computer application program. This chapter also provides some guidelines for writing
applications.
e Chapter 4,

‘“‘Character

Encoding

and

Setup

Escape

Sequences,”

describes the ASCII character. sets and some escape sequences that
initialize and control the DECtalk environment.

e Chapter 5, “Maintenance and Test Commands,” describes the maintenance and test commands used to identify and test DECtalk.
e Chapter 6, “Telephone Communications,” describes how the DECtalk

DTCO03 works when connected to a telephone network.

Xiv

INTRODUCTION

Part 3

DECtalk Phonemics and Voices

Chapter 7, “DECtalk Phonemic Input,” describes the sound system of
English. This chapter shows how to correct DECtalk’s pronunciation to
produce natural, human-sounding speech.

Chapter 8, ‘“Voice Commands and the User Dictionary,” describes how to
use voice commands from the host computer to send phonemic text to
DECtalk. This chapter also describes how to load the user dictionary.
Chapter 9, “Modifying the Voices,” shows how to change the voices
provided by DECtalk and how to create a new voice. This chapter
includes commands to change the speaking rate.

Chapter 10, “How to Get the Most Out of DECtalk,” describes some
techniques for writing applications. This chapter also gives programming
and operating hints to optimize DECtalk performance.
Appendices

The appendices summarize DECtalk’s escape sequences and parame-

ters, phonemic alphabet, abbreviations, acronyms, and homographs. The

appendices also contain a text tuning example, a BASIC program sample,
a configuration procedure, a description of Digital Equipment Corporation’s services, and a list of DECtalk documents.

'GETTING TO KNOW THE SYSTEM

This chapter provides a general description of the DECtalk DTCO03 system. It
covers the DECtalk module and its configuration, controls, diagnostic tests,
and operating mode.
1.1

THE DECtalk DTC03 SYSTEM

The DECtalk DTCO03 system, or DTC03-AA, is intended for use in voice communication systems that have multiple users. Such high-volume systems require a

multiline configuration, as shown with the DTCO03-AA in Figure 1-1. The host
computer sends text to the DTC03-AA, and the resulting speech can be heard

through a telephone.

DTCO3-AA

TELEPHONE
HOST

COMPUTER

NETWORK

SPEECH

“&’" “w\‘w\

MA-1264-85

Figure 1-1

DECtalk DTCO03 System Configuration

GETTING TO KNOW THE SYSTEM

P
:

The DTCO03-AA is a set of text-to-speech converters. These converters are
single-board modules that mount in a rack enclosure. The enclosure contains
eight modules and fits into any standard computer cabinet that is 48.26 centimeters (19 iénches) wide.

1.2

THE DECtaflk MODULE

The DECtalk module, or C5005, is a unit for convartmg text to speech. Each of

o
B

the eight modules in the DTC03-AA supports one telephone line. The modules
connect to one or more host computers through EIA RS232-C serial lines. For

example, Figure 1-2 shows the cabling from the rack enclosure to two host
computers and the telephone network.

The modules receive power through a common power supply in the rack
of the emcimsum powers up all
enclosure. The power switch on the rear paml
eight modul% (Figure 1-2).

DTCO3-AA
E!
RACK

CK ENCLOSURE

POWER

SWITCH

EIA CONNECTIONS

(MODULE SLOTS , T
J1 AND J3)

TO HOST

COMPUTER 1

CABLE MUST

BE SUPPLIED
BY THE USER

_TO FCC/RJ21X OR

DOC/CA21A SERVICE

TO HOST

COMPUTER 2

Figure 1-2
te}

TO LINES 1 AND 3

DECtalk DTC03 Communications Cabling

MA-1266-85

CHAPTER 1

1.2.1

Module Configuration

The DECtalk modules are configured at system installation by Field Service of
Digital Equipment Corporation. Module configuration is done according to the
communication requirements of the user.
Each module is individually configured through an 8-position DIP switchpack.
At power-up, the module controller consults the DIP switchpack to determine

the required baud rate and operating characteristics.
NOTE:

Switch functions and settings are described in Appendix B of this

manual.

1.2.2

Module Controls

To operate and test the DECtalk DTCO03 system, you need to use the controls

on the modules inside the rack enclosure. You can reach the modules by

opening the front door of the enclosure (Figure 1-3).

MA.-0368-85

Figure 1-3

Accessing the DECtalk Modules

GETTING TO KNOW THE SYSTEM

Each module has four light emitting diodes (LEDs) and six pushbuttons on the
front edge (Fegum 1-4). These contm&satl&w ymuto work with individual mod-

ules in the s#yst@m

LEDS

= RING INDICATE
OH | = OFF HOOK

|®©DM] = gusyour
| |©® 0oL | = pata LooP

OL LOOP
~_1l|@®cL | =CcONTR
ehpTEST

PUSHBUTTONS 1 | |

= EXTRA 1

= EXTRA 2

MA-Q364-85

Figure 1-4 Module LEDs and Pushbuttons

CHAPTER 1

1.2.2.1

LEDs - When the module first powers up, all four LEDs light for about

0.5 seconds. This display allows you to detect malfunction in the LEDs themselves. Then all four LEDs turn off while the module runs the power-up selftest.

At the end of the self-test, the LEDs display two types of information: status or
diagnostic.

The following list describes the status displays. When the LEDs display status
information, they light continuously. Each LED represents a single condition.
Most display combinations are valid.

LED

Function

This LED lights when the module successfully passes the
self-test and is ready for operation.

oy,

This LED lights when the module is speaking. During tone

dialing, this LED flashes. Otherwise, it is off.
RI

This LED lights when the telephone is ringing. Otherwise, this

LED is always off.
OH

This LED lights when the telephone is off the hook. This LED

flashes during pulse dialing. Otherwise, it is off.
To display diagnostic information, these LEDs flash an error code or they all
stay off. A diagnostic display indicates module failure. If the LEDs flash or stay
off, call Digital's Field Service for help. See Appendix H for a list of telephone
numbers.

GETTING TO KNOW THE SYSTEM

the pushbuttons activate functions
1.2.2.2 Pushbuttons - When pressed,
such as ,:i‘ag&norstic and loopback tests for checking the operation of the module. When released, the pushbuttons deactivate the functions.
Pushbutton

Function

This pushbutton causes the module to pass over (or “busy
out”) the telephone line. This function can be enabled at
installation if busyout is pmmm@d by the local communica|

tion regulations.

WARNING: In many areas, th@ busyout capability is prohibitedfl. Check with

the local telephone company and call Digital’s Field Service to determine
whether busyout can be enabled in your system.

This pushbutton loops back the trammit data through the

~Thss pushbutton loops back mmd@m control commands

This pushbutton actuvaws the dmgmatw tests, which include
the power-up self-test plus the external data and control

host 1/O port.,

through the hast 1/O port.

loopback tests.

This pushbutton is reserved for use with an optional module.

X1
X2

;

This pushbutton is reserved for use with an optional module.

CHAPTER 1

1.3

THE DECtalk DIAGNOSTIC TESTS

When you turn on the power switch on the rack enclosure, the DECtalk DTC03

system performs a set of self-tests. The power supply and modules are automatically checked for problems.
s,

Pressing the DL, CL, and ST pushbuttons on the modules also activates the
self-tests. You can use these pushbuttons to run routine self-tests on individual

modules.

The results of the tests are displayed on the module LEDs. The next two
sections explain how to use the tests and how to interpret the results.

1.3.1

Power-Up Self-Test

At power-up, the self-test first checks the power supply. If no LEDs on the
modules are on, check the power cord and the ac outlet. If they are fine, the
power supply has failed. Call Digital’s Field Service.

Then, if the LEDs are on, each module runs the test that completely checks the
module logic and 1/O ports. The test runs for about 2 seconds if the communi-

cations line is set at 9600 baud, or for about 30 seconds if the communications
line is set for 110 baud. The test must end with the OK LED on continuously
and the other three LEDs off. At this point, the module is ready to operate.

If the OK LED and any other LEDs are flashing, or if all the LEDs are off, the
module has failed the self-test and must be replaced. Call Digital's Field

Service.
1.3.2

Routine Self-Test

You can run a routine self-test on individual modules anytime during operation.
The routine self-test checks the same things the power-up self-test checks,
except the power supply.

GETTING TO KNOW THE SYSTEM

To run the routine self-test, follow these steps.

1. Open the front door of the rack enclosure.
o

2. On the module you are chackmg, pmss the DL, CL and ST pushbufloms
in that order.

The LEDs on the module light for about 0.5 seconds, then go off for 1

second, and the SP LED flashes a few times.

If the module passes the mst, the OK LED lights continuously.

If the module fails the test the LEDs flash or stay off. Call Digital's Field
Semce

3. The mmdut@ repeats this test every 15 seconds as long as the ST pushbutton is pressed down. To stop the test, release the ST, DL, and CL
pasms the test, it comes up
pushbu&mns in that order. If the madum
-~
-

ready to operat@
NOTE:

Remember to release the ST, DL, and CL
p
PUWWWM or

will continue to run the self-te
st.

~
-~

the module

1.4 OPERATING MODE

to operate when its OK LED lights continuously.
le
is ready
The DECtalk modu
by the host computer and is in operating
controlled
In this state, DECtalk is
mode.

When in operating mode, DECtalk is ready to process input commands from a
telephone and control commands and ASCII text from the host computer.
DECtalk can generate output speech transmission m the user. Chapter 2
describes how thisprocess works.

HOW DECtalk WORKS

This chapter describes how DECtalk converts ASCII data into voice output.

2.1

CONVERTING TEXT TO SPEECH

You enter text and commands into a host computer by using a terminal. The
host computer can then send this ASCII text to DECtalk through the 1/O port.

DECtalk converts this data into speech by a three-level process.

Level1

First, DECtalk accepts text from the host computer and converts
the text from one code into another. The text is in ASCII format
when it enters DECtalk, and is converted to phonemic code for

further processing.

Phonemic code uses the phonemic alphabet described in Chap-

ter 7. Each symbol in the phonemic alphabet has only one pronunciation. DECtalk uses an internal dictionary and the rules of

English pronunciation to do the conversion.

Level
2

Next, the phonemic code is converted into synthesizer control
commands. These continuous variables control the pitch and

timbre for the DECtalk voice.
Level
3

Finally, the speech synthesizer uses the control commands to
generate a speech waveform.

In levels 2 and 3, a synthesizer control command is generated every 6.4
milliseconds, and the digital signal processor generates a speech waveform
value every 100 microseconds. This process generates ‘‘frames” of speech.
DECtalk acts somewhat like a TV picture, in that these frames of speech are
presented to the listener just as frames of pictures are presented to the viewer.

In both cases, the frames appear to be one continuous, unbroken sequence.

HOW DECtalk WORKS

2.2 BECWR SOFTWARE PROGRAM
The thmwlavol process described in the last sectnon happens in seven program module s (Figure 2-1) Each module is des cribed briefly in the following
paragraphs.

C
=

SEPARATES CLAUSES BY PUNCTUATION.
PROCESSES ONE CLAUSE AT A TIME.

PARSER

WORD PARSER AND

| _CONVERTS DIGIT STRINGS TO WORDS.

SF’ELLSQUT UNPRGNOUNCEAELE SVMW’JL STNNGS

NUMBEH FGRMAT‘TER

_'a&mwss FOR A PHGNEM!C Hemfisamm ON

mmm

'FOR EACH WORDPLUS SYNTACTIC INFORMATION.

MANAGER

0 <

WORD

FOUND

| ASSIGNS PHONEMIC
| LeTTER-TO-SOUND
REPRESENTATION FOR

| convERsiON RULE
S|

} EACH uwmwmm wwm

()

PHRASE

| rronunciaTION

|_RuULES

| COMPUTES DURATIONS, PITCH GESTURES, AND ALLOPHONES.
PROCESSES SYNTHESIZER CONTROL PARAMETERS

PHONEME-TO-VOICE |
‘

COMMAND RULES-

@ | DIGITAL SPEECH
|

svsTHESIZER

COMPUTES 10000 WAVEFORM SAMPLES PER SECOND.

T SPEECH

~ WAVEFORM _

MA.-0346-85

Text-to-Speech Conversion

CHAPTER 2

Converting ASCII Text to Phonemic Code
1. A sentence parser breaks the input stream into separate words and
locates some clause boundaries (indicated by commas and other punctuation marks). The sentence parser also recognizes and deals with phone-

mic symbols and commands that you may have added to the input text.
Phonemics are discussed in Chapters 7 and 9.

A word parser breaks words into their component parts, yielding words in
their final pronounceable form. Strings of text that do not form pronounceable English words are spelled out letter by letter.

A number formatter is used if the text contains numerals. The number
formatter knows the rules for many common number formats and converts the numbers into English words. The number formatter also recog-

nizes many common abbreviations, such as “Ib.” for “pound(s).” See
Appendix C for more information on number-speaking rules.
. A dictionary manager searches the pronunciation dictionaries. DECtalk

has a built-in dictionary of many commonly used words. In some cases, a
single spelling has two different pronunciations and DECtalk gives you a
way to indicate which one to use in a particular sentence. See Chapter 7

and Appendix D for the DECtalk pronunciation rules and homographs.
DECtalk also has a user dictionary that can be loaded with words specific

to an application. The user dictionary and its loading are described in
Chapter 8.
A letter-to-sound module uses a set of English pronunciation rules to
assign phonemic form and lexical stress patterns to words not found in

the dictionary.

See Chapter 7 for ways to modify the phonemic form of words, and
Chapter 9 for special voice qualities (such as emphasis and singing).
It is a good idea to program the host computer to automatically send
DECtalk the correct phonemics for the mispronounced which are words

important for a particular application.

HOW DECtalk WORKS

5. A phrase structure module recombines all phonemic output from the
dictionary search and other modules. Durations of phonemes and pitch
commands are computed for the clause, and an appropriate allophone is
selected for those phonemes that éan be pronounced in more than one
way.

Converting Phonemic Code to Synthesizer Control Commands

6. The phoneme-to-voice module processes clauses passed from the
phrase structure module and converts them to control signals for the
speech synthesizer.

This module modifies the clauses by changing the phonemes/allophones
into parameters that determine the natural resonant frequencies of the
vocal tract (formants), sound source amplitudes, and the like.

" The control parameters are sent to the speech synthesizer for output.
Converting Control Commands to Speech

7. The digital speech synthesizer computes a speech waveform with acoustic characteristics that are determined by the synthesizer control commands received.

iy,

COMMUNICATING WITH DECtalk

This chapter gives you an overview of DECtalk programming technique. It
describes the types of data and commands DECtalk uses and how DECtalk

communicates with a host computer.
3.1

PROGRAMMING GUIDELINES

DECtalk is an intelligent peripheral device, so the following guidelines apply.
e DECtalk is easy to control, because the internal DECtalk processor is

sophisticated enough to perform complex operations with simple

commands.

e You can select DECtalk’s operating characteristics and have DECtalk
answer questions about

its status from the host computer. DECtalk can

also inform the host computer of status changes. For example, DECtalk
can tell the host computer if a connected telephone has been answered.
e DECtalk memory stores a substantial amount of useful information. For

example, DECtalk has an extensive built-in pronunciation dictionary. You
can also load a user-defined dictionary.

The following sections describe how DECtalk sends and receives information
from the host computer. Figure 3-1 shows how DECtalk and a host computer
communicate.

COMMUNICATING WITH DECtalk

UPTO8

TELEPHONE

LINES

MA-0351.85

Figure 3-1

DECtalk-to-Host Communications

CHAPTER 3

3.1.1

Types of Data

DECtalk can receive two types of data through its communications connector:
text and commands.

Text is data that DECtalk will speak. Text consists of English-language
sentences, phonemically spelled text, or a combination of both.

Commands are instructions to perform an action. Commands are not spoken

by DECtalk.
You can send commands by using escape sequences, control sequences
(CSI), and device control strings (DCS).

An escape sequence, CSlI, or DCS starts with an ESC character, followed by a
string of ASCII characters. DECtalk interprets the string as a special command.

Character encoding and control functions are described in Chapter 4 of this
manual.

3.2 USING ESCAPE SEQUENCES AND CONTROL CHARACTERS
This section describes the general syntax of escape and control sequences,
and how to use them with DECtalk.

3.2.1

Escape Sequences

The following command is an example of an escape sequence. You can use

this command to enable or disable speaking of text.
ESC
NOTE:

;

ESC \

Escape sequences in this manual are spaced for clarity only. Spaces

are not part of the actual escape sequences.

You can omit parameters with a value of 0. For example, you could send the
above sequence as

ESC

;

ESC \

COMMUNICATING WITH DECtalk

| e

NOTE:

DECtalk does not send 0 parameters in its reply sequences. Because

DECtalk suppresses parameters with a value of 0, DECtalk would send ESC P ;
31; P3z ESC \ in response to the sequence in Figure 3-2 (assuming P3 is not
0). If P3 is 0, DECtalk would send ESC P ; 31 ; z ESC \.

DECtalk escape sequences have the following characteristics.

1. They begin with an ESC character.

2. The ESC character is followed by ASCIl characters that define the
command.

~
-

3. Every character in the command is important. You must enter the exact

characters shown. For example, in the command above, the semicolons
are part of the command. The letter z is lowercase; an uppercase Z has
no meaning to DECtalk.

4. Some commands must end with a string terminator, ESC \. This manual
includes ESC \ with all commands that require it.
NOTE:

DECtalk ignores invalid sequences and commands.

BEGINNING OF ESCAPE SEQUENCE

DECtalk COMMAND

SET FUNCTION

END OF DECtalk COMMAND

END OF SEQUENCE

l
—r—

ESC

;

PHONEME ON

ESC\
DELIMITER

Figure 3-2

Ma-7591-838

Typical Escape Sequence Format

CHAPTER 3

3.2.2

Escape Sequence Format

The following chapters describe the specific escape sequences used with

DECtalk. This manual includes the following information with all escape
sequences (Figure 3-2).
Mnemonic

ASCII characters
Parameters
Decimal value

Figure 3-3 shows the meaning of each part of a typical escape sequence.

DT_QUERY__REPLY (ESCAPE SEQUENCE MNEMONIC)
A

ESCAPE CHARACTER
ASCIll CHARACTER P

ASCII CHARACTER O

ASCIl CHARACTER ;
ASCII CHARACTER 2

]

ESC

v
;

027

080 048

059

050

ASCII CHARACTER 1 (NOT LOWERCASE L)

l——~——DELIMETEH
|

;

PARAMETER

049 079 "‘"{—122

ESC

(ASCIll REPRESENTATION)

027 092 (DECIMAL VALUE)

INDICATES VARIABLE VALUE
MA-7592-838

Figure 3-3

Escape Sequence Representations

COMMUNICATING WITH DECtalk

The mnemonic is a unique name (such as DT_INDEX) used to identify the
sequences in this manual. Mnemonics do not have any direct programming
significance: that is, DECtalk does not recognize a mnemonic name as a valid
escape sequence. However, when you refer to escape sequences by mnemonic in program documentation and program variables, it simplifies editing and
debugging.

The ASCII oharacters are the actual characters to use. Figures 3-2 and 3-3 give
the examples of escape sequences in ASCII format. The escape character is
represented by ESC in all sequences. The numbers that appear are actual
ASCII characmrs not numeric values.

Paramemrs‘ in escape sequences are variables that can cause different
DECtalk actions, depending on parameter values. |
Parameters are represented in this manual by a capital P followed by a num-

ber. Parameters are always sent to and from DECtalk as a decimal number, in
ASCII format.

An empty parameter is treated like a parameter with a value of 0. The
sequences ESC P ; z and ESC P 0 ; 0 z are identical. DECtalk never sends a 0
parameter. However, the 0 parameters are always shown as explicit zeros in
examples.

This manual lists possible parameter values in tables. Two methods are used
to show parameter values.

1. Usually the ASCII character(s) appears (with the decimal value underneath as a check). Use the ASCII character(s) in the escape sequence.

2. Sometimes only a numeric value appears. You must convert the numeric
value to a sequence of ASCII characters for the escape sequence.

The decima}l value of each escape sequence character appears directly under
the character, so you can verify the sequence characters. Parameters are
marked witb asterisks (***), indicating that the value is variable.

Chapter 4 brovides complete tables of all ASCII characters and their decimal,
octal, and hexadecimal equivalents.

CHAPTER 3
T

3.2.3

Special Characters

Some control characters (such as carriage return and backspace) have special
meanings. Table 3-1

lists the special characters that DECtalk recognizes.

DECtalk also recognizes SS2, SS3, CSI, DCS, and ST control characters.

DECtalk ignores all other control characters. (See Section 4.3.2.)

M%;fi

—

,“», a o

;

DR U - i

.
,

;

L ‘V;\

o ,‘;, Dol

e «g%,fi” e
:

e
e fi
o
.
o

e fi&{cv£

L
v

'vw.‘zxfg,%}& vx;;:‘,?

‘

i
o

‘

Gor

»a»*wt%"fi

i /ggy

e
i “":\.
.

;

- .

b
e
b

é‘g@fl

Vil

i Lgnt,a

o
o

ORI

e
e

G G

—

i M’,r} L

it
.

i 'M"{)

L ,(L,‘,figz\@,,.(

W}g
oot

S
L

“’%.p@%fi;fi
.e
'¢.§ Gt

’
e
e

a5
.

.. i

i
G

;

G
e
i
o

‘

s
G

e
i

%Q%‘W’

o e

cann

e
s
-

e
3

'i%,

4‘

e *L

;

o
e

.
o
:

R
5
BS e

G
e

o
o

¢
:

fii -

,p(gfil&« S

e
i

iB
i

‘

ey
L
3

hlnd b6

G
W

g
e

s
:

LUl

‘

s .»«u%y

...
.

ol
o

‘

L
;«”‘&éfigfim Sl

Gld

e
L
:
i

W‘ %w
.

e
i
.
.

. %W;flg

naimane

. Mg::% L

.
o
ii

W“"M
i f‘@“\%
R

.
S

Inn

L
e

.
oW

s 8

G G
e
A

L
i

]

.
e

. ; Av',w& . o
L

i
e

S
.
S S

;
o
e

ion

' '@%{

L
g
: e'%f@%-tfifisgfi e
.

;

é& B
e

f@?{c.,d'w,../
So

.
e

i
L

‘?{@C“@;dg"

G
o

L
d

r?«I

%W’-’"

s
S

;

p—

Abhnrartar

;

)

? .

wm%
el

w,]

""}fifi’

i
"%fi.fi

. .

coan
.

}{/’

o
e

ARG

;”«%

)

e
3

e
e

. ,A‘%%g'%’, L
A

)

i gt

o %

%@ e

G
S

:/e/ o
A

S
8

S
i e

e
2
TN

e
3
%

,@,
o

.
i
i

o
i

R
i
e
G
s

& «,;%‘){ i

.
.

e
o

s
o
e
-

e
i

.
¢

;

i
.

con
D
G

.,
v

COMMUNICATING WITH DECtalk

3.2.4 Effect of the Backspace (BS) Character
If DECtalk finds the backspace character in a word, it modifies the word
according to the hierarchy of the characters involved, as follows.
1. Letters and digits (most important)
2. Punctuation

3. Underline character

The BS character allows DECtalk to process text containing overstrikes and
underscores.

Here are several examples of DECtalk’s processing (spaced for clarity).
Input

Pronounced as
a
a

a BS _
—_BSa
aBShb
BS de
ab BS

3.3

b
de

DECtalk-COMPUTER COMMUNICATION

Programming DECtalk is similar to programming a smart terminal (such as a

VT220). DECtalk and the host computer must exchange information according

to fixed rules.

DECtalk does not process text until it reaches a clause boundary. Clause
boundaries mark the end of phrases or sentences. DECtalk recognizes the
following clause boundaries.

e A period, comma, exclamation point, and question mark are boundaries.
Certain other punctuation characters, such as ) and (, are also boundaries. If a period is used, DECtalk checks the characters after the period
(because periods do not always mean the end of a sentence).
e A full buffer acts as a boundary. If DECtalk’s temporary buffer begins to
approach its fill limit (at about 32 words), DECtalk begins speaking what is
in the buffer, as if a comma appeared in the text.

e A timeout is another boundary. If nothing is sent to DECtalk for 5
seconds, and there is text in the buffer, then DECtalk speaks all the text in
the buffer as though a comma had been sent with the text.
iy

CHAPTER 3

Escape sequences represent (1) commands sent from the host computer to
DECtalk, and (2) status replies sent from DECtalk to the host computer. All
escape sequences begin with the ESC character; a sequence ends when the

last character required for that sequence is sent. Do not use a carriage return
or any other normal terminating character to terminate an escape sequence.
Figure 3-4 shows the data paths in DECtalk, as follows.

1. The DECtalk unit is in the center of the figure. The speech processor is

part of the DECtalk unit, but is shown as a separate module.
)

—

e YL

—

2. Arrows show the direction of information flow. Notice that information
from the telephone flows to the host computer and from the host computer to the speech processor. Information from the speech processor is
sent to the telephone.

3. Each DECtalk escape sequence affects the flow of information within
particular data paths. The switches within the data paths represent the

points at which the escape sequences act. For example, the DT_STOP

—

escape sequence affects the data flow from the host computer to the

speech processor.

4. Because there are a large number of commands and parameters, they
are grouped in boxes under the figure.

You can control and use DECtalk in many ways when it is connected to a host
computer. The rest of this chapter, Sections 10.1 and 10.2, and the sample
program in Appendix E describe a general programming method for DECtalk. If
you are writing a control program for DECtalk, remember that your application

and needs may not exactly match the descriptions that follow.

—

COMMUNICATING WITH DECtalk

M "?tm%
TELEPHONE =
eed SPEECH
PROCESSOR | 170
N«
|i m -

‘

HOST SPEAK
M._.___

I"‘l " %m‘m‘

TERMINAL

HOST
COMPUTER

~
'
—O—0O— SWITCH ON

DECtalk
'

—0— 0— SWITCH OFF

SPEECH PROCESSOR SEQUENCES

TELEPHONE COMMUNICATIONS

DT__MODE:MODE__SQUARE
DT_MODE:MODE__MINUS
DT__MODE:MODE__EUROPE

| DT__PHONE
DT__MASK
—

DT__MODE:MODE__SPELL

DT__PHOTEXT

HOST SPEAK SEQUENCES
DT__STOP
DT__SPEAK
DT__SYNC

MA-0364-85

Figure 3-4 DECtalk and Host Computer Programming Interaction

CHAPTER 3

3.3.1

DECtalk Setups

Each module in the DECtalk DTCO03 system must be individually set up, as

described in Appendix B. The following events may occur at power-up.
1.

Programs may send a ‘““What are you?” sequence to make sure DECtalk
is available. DECtalk replies with a certain ID code to identify itself.

. The device attribute request (Section 5.2) describes the ‘“What are you?”
sequence.

. Set MODE_SQUARE on. This command ensures that phonemic code

values are accepted. See the DT_MODE command in Section 4.6.2.
If you connect DECtalk to the public telephone network, select the correct
telephone handling parameters. See Chapter 6 for these parameters.
You may have to set up the host computer (or the DECtalk communication

line) for DECtalk commands. You must set up the computer for singlecharacter,

type-ahead

input,

and

operating

system

XON/XOFF

processing.
i

If your host computer cannot support single-character processing, you
can use the DT_MASK command to permit line-at-a-time processing. See

Section 6.2 for information on the keypad mask command (DT_MASK).
o,

Other commands depend on the DECtalk environment, such as special
text-to-speech commands.

3.3.2

Program Control

DECtalk is primarily a speech output device. It assumes the host computer will
handle most of the necessary control operations (such as waiting for task

completion and requesting status).

COMMUNICATING WITH DECtalk

3.3.3

Data Synchronization

DECtalk’s speech rate is much lower than the potential data transfer rate on
the host communication line. A standard XON/XOFF flow control protocol is
used to prevent the overflow of an input buffer on each module.

All received characters are stored in a 64-byte input buffer. DECtalk sends an
XOFF character to the host computer when its input buffer is filled to a turn-off
limit (32 characters), asking the host computer to stop sending data. Meanwhile, DECtalk continues to process the characters from the buffer. When
DECtalk’s input buffer empties to a turn-on level (16 characters), it sends an
XON character asking the host computer to start sending data again.

DECtalk’s input buffer is large enough so that the host computer can continue
sending data at the highest speed (9600 baud) for up to 250 milliseconds after
it receives the XOFF, without losing the data.

Figure 3-5 shows how DECtalk synchronizes data transfer with the host computer through XON/XOFF signals.

If the host computer does not stop sending data in time, the input buffer may

overflow and characters may be lost. DECtalk does not give an audible warning

of this overflow, except for the obvious garbling of partial words. The host
computer can issue a device status request (DSR) command to determine if an
input buffer overflow occurred.

Most operating systems have a HOSTSYNC option (or its equivalent) in their

terminal characteristics. If this characteristic is set on the DECtalk communications line, the host computer handles XON and XOFF characters. If XON/XOFF
coordination is not available, you may be able to avoid buffer overflow by using

the DTMSY?NC command to control the application program’s output rate. However, Digital does not recommend this method because it is difficult and errorprone. Section 8.6 discusses the DT_SYNC command.

CHAPTER 3

HELLO. | AM DECtalk. WHEN IN
THE COURSE OF HUMAN
EVENTS.... MUMBLE.... MUMBLE...

...OF HUMAN

EVENTS...

..WHEREFORE HE HATH CAUSED.
"

..MUMBLE .. . MUMBLE...

—

XOFF

HOST COMPUTER PAUSES WHILE DECtalk

omgy

EMPTIES ITS BUFFER.

XON

| DECtalk’'s BUFFER IS NEARLY EMPTY, SO IT
SENDS XON.

[ -1 THANK YoOU FOH...]

N
..MUMBLE . . . MUMBLE...

I THANK YOU FOR

LISTENING TO ME.

HOST COMPUTER RECEIVES XON AND RESUMES TRANSMISSION.

MA-0349-85

Figure 3-5

Synchronizing DECtalk and Host Communications

CHARACTER ENCODING AND
SETUP ESCAPE SEQUENCES

DECtalk has several parameters that can be changed to conform to the operating environment. These parameters include the following.
e Line characteristics (such as line speed)

e Character sets (to send and receive information)
e Modes (to control DECtalk’s interpretation of special characters and phonemic text)

There are also several testing and inquiry commands, described in Chapter 5.

4.1 SELECTING ASCIl CHARACTER SETS
DECtalk is a computer terminal device, and conforms to the standards for
computer terminals.

DECtalk uses the following rules for all character sets.

1. Upperbase and lowercase letters are considered the same (except for
some abbreviations that are dastmgunshed by case). For example, DECtalk
speaks the letter G as ‘‘gee,” not ‘‘uppercase gee.” (See Appendix C.)
2. When found in words, foreign letters (as found in the multinational character set) are spoken as in English. For example, DECtalk speaks the letter
a as “‘ey umlaut.”

CHAPTER 4

3. DECtalk operates with 7-bit and 8-bit data. You can translate or map 7-bit
codes into 8-bit codes, and 8-bit codes into 7-bit codes
mapping has no effect on spoken text. Table 4-1

(Figure 4-1). This

gives the escape

sequences that change DECtalk to 7-bit or 8-bit modes.

P MAPS INTO ITS

8-BIT EQUIVALENT 208.

BITS
&4 83 82 81
v

[

)

[

COLUMN

[

\ 00

| bl

ML § DLE

[

[1 1

]

’M gansnss] r_____

w16 | SUB

M
R

ROW

]
°

[

LAl

1320

e
ESC MAPS INTO ITS
8-BIT EQUIVALENT CSI.

MA-7594-83

Mapping 7-Bit and 8-Bit Tables

e L

AP,

Figure 4-1

Ay,

e
o

oi
A"

v1 i

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

CODIMG STANDARDS

”

4.2

technique that are compatible with the following ANSI and ISO standards. ANSI

(American National Standards Institute) and ISO (International Organization for
Standardization) specify the current standards for character encoding used in

the communications industry.

Standard

Description

ANSI X3.4 - 1977

American Standard Code for Information

DECtalk uses an 8-bit character encoding scheme and a 7~bnt code extension

Interchange (ASCII)

ISO 646 - 1977

7-Bit Coded Character Set for
Information Processing Interchange

ANSI X3.41 - 1974

Code Extension Techniques for Use
with the 7-Bit Coded Character Set
of American Standard Code Information
Interchange

ISO Draft
International

Standard 2022.2

ANSI X3.32 - 1973

7-Bit and 8-Bit Coded Character
Sets - Code Extension Techniques

Graphic Representation of the

Control Characters of American

National Code for Information

Interchange

ANSI X3.64 - 1979
'

Additional Controls for Use with
American National Standard for

Information Interchange

ISO Draft

Additional Control Functions for

International

Character Imaging Devices

Standard 6429.2

4.3

CODE TABLES

A code table is a convenient way to represent 7-bit and 8-bit characters,
because you can see groupings of characters and their relative codes clearly.

CHAPTER 4

4.3.1

7-Bit ASCII Code Table

Figure 4-2 is the 7-bit ASCII code table. There are 128 positions corresponding

to 128 character codes arranged in a matrix of 8 columns and 16 rows.
gy

COLUMN

ROW|

BITS

40
SP | 32

60
48

100
64

120
80

140
%|

160
|2

101

121

141

102

122

142

Oloooo

0
o

20
©]
21

|NUL|

0
1

2l
3l

ooo1

o0 10
0011

o100

0101

|SOH|

|17

|STX|

2 | DC2]|

JETX |

111
121

[

103

123

143

124

144

164

105

125

145

5 |NAK

6
7

7
7

26
27

|CAN|

14|

1 10 | SO

18]

23
2

107

40
28
51

a1
29

110

111

57
39

13l
a9

| e
55

126

| s

127

87
57

130

88
58

116
74

146

147

03]

150

104
68

131

151

89
59

w0s|
69

165

e
75

166

| s

167

w | 119

170

120
78

171

121
79

|42

58 |

112

132

152 |

133

163

173

27|

+ | a3

|ESC|

26|

106

|SUB|

)

26
39

(

]

|10

|2 | FS | 28

CR | 3|

16
|
36
|14 | RS | 30|

E
17

US | =

2C
55

1E
37

| 5|

g | 32

GS | 2

|ETB|

% | 37

|[SYN|22|

HT | 9o | EM | 5
9
19
12

114

[
\

| %

58
134

107
68

154

o8]

135

156

56
a6

76
62

116
|

aE
17

136
94

109

156
110

{
‘

172

| 122

123

174

124
7C

}

175

125

176
126

137

157

11|

DEL | 127

115

163

104

r | 14

| LF

162

[|DC4 | 2

|ENQ|

161

cHaracTer|

131

"ol

|EOT|

1000 )| BS

1+ o 1 1

|BEL|

0111

oz | 19

3
33

10]

|]ACK|6s

1 001

o110

DC1

' | 5o

{ DLE|

b4 b3 b2 b1

7€
177

OTM

ARG,

| pecimAL

§ HEX
g,

MA-0B893A-83

Figure 4-2 7-Bit ASCII Code Table
32

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

Each row represents a possible value of the four least significant bits of a 7-bit

code (Figure 4-3). Each column represents a possible value of the three most
significant bits.

Figure 4-2 shows the octal, decimal, and hexadecimal code for each ASCII
character. You can also represent any character by its position in the table. For

example, the character H (column 4, row 8) can be represented as 4/8.
DECtalk processes received characters based on two character types defined

by ANSI: graphic characters and control characters.
Graphic characters are the characters that can be displayed on an output

device (video terminal or hardcopy printer). The ASCII graphic characters are in
positions 2/1 through 7/14 of Figure 4-2. They include alphanumeric characters

plus punctuation marks and various text symbols. Examples are C, n, ", !, +,
and $.

Control characters are single-byte codes that perform specific functions in data

communications and text processing. The ASCII control characters are in positions 0/0 through 1/15 (columns 0 and 1) of Figure 4-2. Most control characters
are silently ignored. The SP character (space, 2/0) can be considered either a
graphic character or a control character depending on the context. DEL (7/15)
is always used as a control character.

Control character codes and functions are standardized by ANSI. Examples of
ASCII control characters with their ANSI-standard mnemonics are CR (carriage

return), FF (form feed), and CAN (cancel).

BIT
7

BIT
6

3mosT

BIT
5

BIT
4

SIGNIFICANT BITS

BIT
3

BIT
2

BIT
1

4 LEAST

SIGNIFICANT BITS

BIT
8

»|

BIT
7

BIT
6

BIT
5

BIT
4

BIT
3

»le

4 MOST

SIGNIFICANT BITS

BIT
2

SIGNIFICANT BITS

(DECIMAL VALUE IS

(DECIMAL VALUE

(DECIMAL VALUE IS

IS ROW IN

COLUMN IN

IS ROW IN

CODE TABLE)

Figure 4-3

4 LEAST

COLUMN IN

7-Bit Code

BIT
1

(DECIMAL VALUE

8-Bit Code

MA-0891-83

MA-089083

Data Word Representation

CHAPTER 4

4.3.2

8-Bit Code Table

The above conventions can be generalized to the 8-bit character encoding

used on DECtalk. Figure 4-4 shows the 8-bit code table. It has twice as many
columns as the 7-bit table, because it contains 256 versus 128 code values.

COLUMN

\ 00 | 01 | 02 | 03 | 04| 05|06} 07 08|09} 10 1112 | 13 | 14 | 15
ROW

ey,

NUL | DLE | sp

SOH | DC1

PU1

sTX | DC2

PU2

ocs | /i

liony

ETX | DC3

STS

EOT | DC4

IND | CCH

ENQ| NAK

NEL | MW

ACK | SYN

SSA | SPA

| BEL | ETB

ESA | EPA

BS | CAN

HTS

HT | EM

HTJ

LF | suB

VTS |

VT | ESC

PLD | CSI

FF | FS

PLU | ST

CR | GS

sO | RS

ss2 | PM

si |

| OSC

DeL | ss3 | APC

Moo
copesTMf—
Gt CODES

~t<crcopesTTM

7-8IT CODE TABLE

GR CODES

MA.-0892-83

Figure 4-4

8-Bit Code Table

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

As with the 7-bit table, each row represents a possible value of the four least
significant bits of an 8-bit code (Figure 4-3). Each column represents a possible
value of the four most significant bits.

All codes on the left half of the 8-bit table (columns 0 through 7) are 7-bit
compatible: their eighth bit is not set and can be ignored or assumed to be 0.
You can use these codes in either a 7-bit or an 8-bit environment. All codes on
the right half of the table (columns 8 through 15) have their eighth bit set. You
can use these codes only in an 8-bit environment.

The 8-bit code table has two sets of control characters, CO (control zero) and
C1 (control one). The table also has two sets of graphic characters, GL (graphic
left) and GR (graphic right).

The basic functions of the CO and C1 codes are as defined by ANSI. CO codes
represent the ASCII control characters described earlier. The CO codes are 7bit compatible. The C1 codes represent 8-bit control characters that let you
perform more functions than those possible with the CO codes. C1 codes can
be used directly only in an 8-bit environment. Some C1 code positions are left
blank because their functions are not yet standardized.

7
.

‘In Figure 4-4, all control characters that are recognized by DECtalk are highllighted. All other controls are ignored.

The GL and GR sets of codes are reserved for graphic characters. There are

94 GL codes in positions 2/1 through 7/14 and 94 GR codesin positions 10/1
through 15/14. By ANSI standards, positions 10/0 and 15/15 are not used. You
can use GL codes in 7-bit or 8-bit environments. You can use GR codes only in
an 8-bit environment.

4.4

CHARACTER SETS

You cannot change the functions of the CO or C1 codes. However, you can
map different sets of graphic characters into the GL and GR areas of the code
table.

DECtalk supports the U.S. ASCIl (ASCII_G), DEC Supplemental, and U.K.
National character sets. Figures 4-5, 4-6, and 4-7 are the code tables for the
ASCII_G, DEC Supplemental, and U.K. National character sets respectively.
These character sets are the only sets understood by DECtalk.

CHAPTER 4

COLUMN

b7
-

ROW|

b4 b3 b2 b1

O|loooo

0
0

0001

0011

)
10

11]
12]

+
1

13}

0o 1
1

1
0

18]

142

161

|13

123

143

163

19 |

‘64

104

124

144

6
6
7

(

29
36

|s3

|54
36

|5}

39
72

3B
74

“ul

2C
55

)

D | s

|s2
34

| 3

46
38|
26

%
&

22|
16

3c
75

62]

125

145

101

106

126

107

127

110

130

150

| 7w
46
7

86
56

146

102]
66

147

131

151

05|

112

132

152

106

113

133

153

114

134

154

o8|

135

116

136

107

155

109
6D

156

110

114

116
74

165

166

| s
76

| 119

120

167

77,
170

172
z

{

173

123
174

124
7C

}

175

125

176

126

117

137

157

177

DEL | 127

122

oy,

171

iy,

|17

164

| s

162

wo |

105

99|

122

103

102

o 1
0

14]

141

121

101

| s

160

|12
70

10]

%|
60

140

80
50

120

1000

0
100

@ | 64
40

|[4]
30

0111

1
t

o110

0
[ 40

1
1

SP | 2|
20

o100

7€

ls——C0 CODES—

GL CODES

(ASCII GRAPHICS)
R,

cHaracTer|

EgC | 3 | O°TAt
27

| DECIMAL

HEX
MA-0893-83C

Figure 4-5

ASCII_G Character Set

Ay,

oo,

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

8
1

9
'

10
1

0
o

12
1

15
1

T
1

COLUMN
o]

200

220 V //}

128
80

| 144 §
90

201

221

129

s |

202

222

130

we|

/7]

|
¢

240

160
Ao

241

w | 4+
Al

242

|82

223

243

w| £

203

131

|ie3]

|20

176
oj

261

|302

[263]

[303

244

64|

225

133

164

wel

245

fws)]

224

205

180

.,
|

265
|w]

206

226

246

266

a6 |

86 |

134

150 |

166

207
135

227
w|

210
136
88
211

230
2|
98
231

w3l
99

212

232

97|

137
89

138

154

213

139

185 1

%7 )
|w]

250
|es
Ag |
251

270
wa|l
B8 |
|an}

252

272

|69
A9

170

233

263

[

214

234

254

140

156

72|

215

235

256

141

157

216
182
8E
217

236
158
9E
237

256 |
174
AE
w1

143

159

l‘"——C1 1 CODES

s ]

::L:

1821

247
|wer|]

ws|
B9

|wes]

|30a

§
§

|35

|w]

|307
|w|

o
g

|310
|20
c8
|3

| 314

275

03]

|204]

315

051

276
wo|
BE
|27

|38
|08
CE
|3V

|19t

oe
|

|325
|23]

200

a
a8

327
|25]

330
26|
D8
331

|27]
D9

|28

|29

334

(20|

335

&
®

336
222
DE
337

A
P

|223)]

,
(DEC SUPPLEMENTAL
GRAPHICS)
GR CODES

214

[333

332

’

a2 |

|313]

(2ot
co

202

RET

312

|we]

o
@

|20

323

|32

[s)

[a87

186

274 |

|200

306 | .,
198

,
g

|32

322

196

208
DO

273 |

Bal

320

192
co

301 |

148

[w2]

204

132

A
T

[300

60 |

224
EO

240}
FO

341

|25|

342

|28]
||

f3aa]

|2e2]

|23]

|36

|as |

|244a]

|38

|20

363

|28]

a6}

[2]

|32

|29

361

343

|25

347
|221]

350
[22|
E8
351

||
E9

352

”

000

1001

|29
Fa

[]

®s |

239

374

|z2]

|35

¥V

|38
| 238
EE
w1 77

0 1 0110

01 00 | 4

372

|[23]

376
40
FE
) a7
12551

(0]

|27]

o011

|33

001 0|2

0111

370
|28]
F8
3n

367
|47

|25

o001

234

354

0o 0010

01101}

w3l

JROW

261

|366

b4 b3 b2 b1

0101

»7 BITS
b6

10110

11 00

|12

v+ 101

|13,

1110

|14

|15

111

-|
MA-0894-83C

Figure 4-6

DEC Supplemental Character Set

CHAPTER 4

COLUMN

BITS

———

b5
ROW| b4 b3 b2 b1
Oloooo
11
2]l

0001
00

0
0

6]
10

1000

13]

o 1

141

18]

141

102

122

142

162

)
23
a4

(3|

46
38

4
85

|
98

(

| s
3 |
64

|52

34
65

|sa|

D
E

7
8

66
|54

36
67

|55
37

7]
43
104

[e]
44

105

|6]

| 84

| s

G
H

46 |
107

125
55

106

nlt

83
53
124

110
]

w
X

126
86

56
127

87
57

130
88

9|l
63 |
144

wo]
64

145

161

|13

.4
72
163

|15
73
164

102

147

03]

150
104

IRTF
75

166

167

w | 19
|

170

120

111

131

151

171

)

2A
53

| a3

9
.
:

2C
55

45
56

]

3A
73

so|

1
J

|[e0]

3C
75

61|

| e2]

63
3F

4A
113

Y
b 4

[

)

4c
115

]

112

114

el
78

n7|

89
59

132

133

106
69

152

106
6A

153

107
68

Yy
z

{

121

172

122

173

123
78

134

154

135

155

156

176

136
94

109 |

110

174

}

Dy,

124
7C

175

125

126

137

157

DEL | 127

TS,

|18

165

146

|16

19
13
24

|2
70

143

3
3
4

160

12]

6|
60

123

o110

o 1

| 140

| e
50

103

22l

||
40

120

o101

|#]
30

1 o

101

- 100

o100

001

00 11

SP | 32|
20
!

0 1

177

CHARACTER

| OCTAL

] becimaL

¥ HEX
MA-0893-83S

Figure 4-7

4.4.1

U.K. National Character Set

Designating Alternate Graphic Sets

DECtalk has four alternate graphic character sets, called GO, G1, G2, and G3.
Before a character set can be invoked (used), it must be designated (loaded)

into one of the four alternate sets.

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

At power-up, DECtalk designates the ASCII_G character set into GO and G1,
and designates the DEC Supplemental character set into G2 and G3. You can
modify these default designations by using the sequences in Table 4- 2. Figures
4-8 and 4-9 show how you can designate the character sets.

»@m,.%:‘

g
e

I e
Dy

S
S

o
:

4
S

e
o

4“{4 L

‘fi i

e
e

=
h:.
=

.
i

s
=

R
e
.

= =

I . 'v»

e
e e

:
.

&w\fi‘f{“‘

Spras

=~
-

Cal

e
i

=
o

.
-

.
s
;

oj
b2

—

{%\t»,
o
i o

N
i
i
L
e
G

CHAPTER 4

]

—

INVOKING

SEQUENCES
(LOCK SHIFT AND

SINGLE SHIFT)

\\\\\
-

-t

G1
[

G3
N

DESIGNATING
SEQUENCES
|

GRAPHIC |
|
REPERTORY
1
MA-Q279A-82

Figure 4-8

Loading 8-Bit Characters
A

ool

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

—

7.81T
SET IN USE

O
+

O+
@

@ sss

*ALWAYS ASCII OR THE CURRENT.
GRAPHIC

REPERTORY|

NRC SET IN VT100 MODE AS
SELECTED IN SETUP.

MA-0280A-82

Figure 4-9

Loading 7-Bit Characters

CHAPTER 4

4.4.2

Invoking Graphic Sets into GL and GR

Character sets that have been designated into one of the four G-sets may be
iy

used by invoking the set into the GL or GR areas of the code table.
G,

At power-up, DECtalk invokes the GO character set (which holds ASCII_G) into

GL, and the G2 character set (which holds DEC Supplemental) into GR. You

can modify these default invocations by using the sequences in Table 4-3.
AR,

’M%(a

AR,

}

e,
2

)

/ jvfi
arosnz,

4.4.3

Character Processing

In 7-bit mode, characters received from the host computer are either CO control

characters or GL graphic characters. All CO control characters are interpreted

as described in Section 3.2.3. All GL graphic characters are processed according to the rules for the character set currently invoked into GL.
SR

In 8-bit mode, characters received from the host computer can also be C1
control characters or GR graphic characters. All C1 control characters are

interpreted as control functions. All GR graphic characters are processed

according to the rules for the character set currently invoked into GR.
At power-up, due to the default designations and invocations,

DECtalk

processes characters in a way that is compatible with Digital’s other terminal

products.

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

4.5 CONTROL FUNCTIONS
You can express control functions as single-byte or multibyte codes.

The single-byte codes are the C0O and C1 control characters. Your program can

perform a limited number of functions with CO characters. C1 characters give
you a few more functions, but your program can use them directly only in an 8bit environment.
Multibyte codes represent far more functions because of the variety of code
combinations possible. These codes are called escape sequences, control
sequences, and device control strings.
4.5.1

Escape Sequences

An escape sequence is a sequence of one or more ASCII graphic characters
preceded by the CO character ESC (027). For example,
ESC

(

027

040

066

is an escape sequence that requests DECtalk to designate the ASCII_G character set into GO.

Because escape sequences use only 7-bit characters, you can use themina7bit or 8-bit environment.

Escape sequences have an important, special use. ANSI permits code extension techniques to extend 7-bit control functions. In particular, you can use
two-byte escape sequences as 7-bit code extensions to express each of the
C1 control codes.
You can express any C1 control character as a two-character escape
sequence in which the second character has a code that is 40 (hexadecimal) or
64 (decimal) less than that of the C1 character. Conversely, you can make any

escape sequence whose second character is in the range 4/0 through 5/15 one
byte shorter (if running in the 8-bit mode) by removing the ESC and adding 40
(hexadecimal) or 64 (decimal) to the code of the second character, thereby
generating an 8-bit control character.

CHAPTER 4

For example, the C1 characters CSI, SS3, and DCS can be expressed as
follows.

C1 Character

7-Bit Extension Equivalent
(escape sequence)

CSli

ESC [

155

027 091

SS3

ESC O

143

027 079

DCS

ESC

144

027 080

4.5.2

Control Sequences

A control sequence is a sequence of one or more ASCII graphic characters
preceded by the CSI symbol (9/11). CSI can also be expressed as the 7-bit

extension code ESC [. For example, the following control sequence requests a

device status report from DECtalk.CSlI

155

053

110

ESC

[

027

091

053

110

A sequence starting with CSl can only be used in an 8-bit environment because

CSl is a C1 control character.

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

Device Control Strings

4.5.3

A device contro string is a delimited string of characters that are used in a data
stream as a logical entity for control purposes. It consists of an opening

delimiter (DCS introducer), a command string (data), and a closing delimiter (a
string terminator or ST).

DCS (9/0)is an 8-bit control character. You can also express it as ESC P when
coding for a 7-bit system. ST (9/12) can also be expmssad in a 7-bit environment as ESC \. For example, the following instruction allows DECtalk to speak
phonemic text.

048

122

048

059

ESC P
027 080

0
048

;
059

0
048

z
122

pcs 0
144

156

ESC \
027 092

4.6 WORKING WITH 7-BIT AND 8-BIT ENVIRONMENTS
DECtalk supports both 7-bit and 8-bit communication techniques. To take

advantage of DECtalk’s 8-bit character set, your program and communication
environment must be 8-bit compatible.
4.6.1

7-Bit and 8-Bit C1 Controls

Communication parameters are set up through two switches on the DECtalk
module. (See Appendix B.) Switches SW4 and SW5 determine if the basic
transmission and reception path is set for 7- or 8-bit communication. Four
operating modes are possible:

1. 7 data bits and even parity
2. 7 data bits and odd parity

3. 7 data bits and ignored parity (8 bit, no parity, and ignored most signifi-

cant bit)

4. 8 data bits with no parity

CHAPTER 4

If one of the first three modes is set, you cannot send 8-bit characters to
DECtalk, and DECtalk will always send C1 control characters in their 7-bit
format.

If the communication parameters are set for 8-bit mode, then full 8-bit format

(controls and data) is allowed. Characters in the C1 and GR sets are treated as

such. You can use special escape sequences (S7C1T and S8C1T) to control
the transmission format of C1 control characters. At power-up, DECtalk simulates the S8C1T sequence.

The S7C1T sequence causes DECtalk to send all C1 control characters in 7-bit

format.

ESC

027

032

070
S

The S8C1T sequence causes DECtalk to send all C1 control characters in 8-bit
format. This sequence is ignored if the transmission and reception path is set

to 7-bit mode.

4.6.2

ESC

027

032

071

Mode Selection (DT_MODE)

The DT_MODE command provides a number of options to control how
DECtalk handles particular characters in spoken text. The general DT_MODE
escape sequence is as follows.

ESC
P

027

080 048

;

059

056

;

050 059

;

ESC
\

***

059

***

122

027

092

The P3 parameter is interpreted as a set of flag bits. P3 uses only four flags
(Table 4-4).
At power-up, MODE_SQUARE is set (that is, the square brackets [ and ] are
used as phonemic delimiters), and all other modes are cleared.

CHARACTER ENCODING AND SETUP ESCAPE SEQUENCES

The P4 parameter controls the interpretation of P3 as follows.

If P4 isiO, then P3 changes the mode.
If P4is

1, then P3 is a set of bits (flags) to be set in the mode.

If P4 is 2, then P3 is a set of bits (flags) to be cleared in the mode.

Loln

;

e
B
o

% e

1' L
S

L
foel

5}%.‘

.
Goadian

%Q L .

i o
SR

‘“s‘f«f:,'f*}‘lw??
i

Lo
i

)

i
i

:’4@'@5@\ .

.
i
’

e
et

o
e

o
o

L
.

i
-

{

=
=-

o
e
;

e
-~.

Ay
o

s. i

""“';‘“Vf{‘g?@3 .

o .
..

.e
)

‘;%Q@

.
o

*‘
.

)

%fi e

v
v

"fi

"%fi’ i

fz)g.;

.
e

- s
.

5:;.= S

—

.
.

G
i

.
i

=—-

i
o

L .

L
.

.
G

é«

i *v«r{:a.fiti:;"y‘

o
L

vmw‘

.
i

o
Wi

‘

;

. \.%'l“é’f:

‘

.
R

e
i
e

i
i

i
sl
i

i
W

—=

i
o

=-o

=
e

;

PSR

-z
—et

= w’gf

..
A«%}flg@fi%‘ i %{%, &ég{fi/\
(,'

- 5“@
.

CHAPTER 4

Use the following method to get the P3 value.
1. Add up the values of the MODE flags in Table 4-4 that you want to use.

2. Convert the sum to ASCII digits. Use these digits in place of P3 in the
escape sequence.

For example, assume you want to set MODE_SQUARE and MODE_MINUS,

and leave the other flags unchanged.
MODE_SQUARE

MODE_MINUS

Desired P3 value =

Enter the P3 value 5, and then the P4 value 1. The resulting escape sequence
is as follows.

ESC
027

;

8§

;

080 048 059 056 050 059

053 059 049

ESC
\

122

027

092

DECtalk will have the MODE_SQUARE and MODE_MINUS features enabled.

MAINTENANCE AND
TEST COMMANDS

DECtalk uses a set of commands that identify its configuration and certain
operating features. These commands also help to test a DECtalk unit remotely.
Most of these commands have an inquiry-response format. DECtalk returns an
answer to the host computer after the action is complete (or in response to a
pure inquiry).

5.1

CONFORMANCE LEVEL REQUEST (DECSCL)

" Device conformance (or compatibility) level sets the level of interface compatibilty and determines the specific functions a device can perform. DECtalk is a
level 2 speech output device that can run in level 1 compatability mode.
You can use the following sequence to set the conformance level.
ESC[
037 091

P1
**

;
059

P2
**

”
034

p
112

The P1 parameter must be eith’er 81 (for level 1) or 82 (for level 2).
The P2 parameter selects the C1 control transmit format. It is either 0 (default
setting, 8 bits), 1 (7 bits), or 2 (8 bits). The 0 and 2 parameters are ignored if the
communication channel is set for 7 bits.

You can find out the DECtalk conformance level by using a DA primary command shown in Section 5.2.1. Because the conformance level affects how text
is processed (in particular, it determines the interpretation of the
MODE_EUROPE and MODE_SPELL commands), the DECSCL command performs an implied DT_SYNC command.

CHAPTER 5

5.2

DEVICE ATTRIBUTE REQUESTS

DECtalk has two commands that may be used for device identification.
5.2.1

Device Attribute Request (DA Primary)

The device attribute request escape sequence is as follows.

ESC

[

027

091

099

DECtalk identifies itself by sending the following status reply sequence.

ESC[
027

091

;

063

056

050

059

050

059

051

099

This DA response provides the device code (82) to indicate that DECtalk is a

level 2 device, and also shows wink detection (code 2) and call progress (code

3) extensions.
5.2.2

Device Attribute Request (DA Secondary)

The secondary DA request escape sequence helps to identify the hardware/firmware configuration of the DECtalk unit.

ESC

027

062

099

DECtalk responds with the following status sequence.

ESC

;

027

062

049

050

059

049

048

059

049

099

The first parameter is the hardware identification code. For DECtalk, this code
is 12.
The second parameter identifies the firmware version number. In this case, the

code 10 indicates version 1.0 of the firmware.

The third parameter is 1 to identify the language spoken by the DECtalk unit as
American English.

MAINTENANCE AND TEST COMMANDS

DEVIC TEST AND STATUS

5.3

A special set of escape sequences run DECtalk hardware self-tests. Another
set of escape sequences forces DECtalk to return status reports. The following
paragraphs describe these sequences.

5.3.1

DEC

Ik Power-Up Status

You can reset DECtalk to its power-up state. The method you use to reset
DECtalk ma

affect the operating features. You can reset DECtalk with any of

the foilowing methods. This section describes methods 2 through 4.

1. Power-up (PUP) is the state that DECtalk is in when first turned on.
2. Return to in itial state

(RIS) is a hard reset you can request by using an

escape sequence.

3. So

reset (DECSTR) is a partial reset you can request by using a control

sequence.

4. Nonvolatile memory reset (DECNVR) is included for compatability with
earlier DECtalk devices.
Table 5-1 lists the DECtalk factory default settings.

Table 5-2 lists some other DECtalk actions performed through listed reset
methods.

(;i-n“
a

)

‘e

‘

S
.

. il

- \w .

dalsiin

e
e

Gl i

"}“»‘»'f"'m&g% s

i
e

L
i
.

L
i

CHAPTER 5
A

. sl st
w?g .

.
o

L
sl
st

‘

SRR

oy
e

L 4:0(.‘"“,'

D "'fl}}‘:’%fi;"

.
Kgfi( e
G

‘

G
i

e
e

.
P

fhati

e g@%@ L

e
i

.
e

:‘ft}

oyl

..
L

e
i

s
B

2
o

»;k«"’k«"“

iy
7

- %}%’2{% "%fi

. :

w'fifi‘« i
-

e
A
o
;.

s
i

)oé;:(:w‘"

i
o

5.3.2

mj?»%@:

o
G

RAY

A,
,,,

Device Self-Tests (DECTST)

You can use this sequence to initiate DECtalk self-tests remotely. Appendix B
shows how to perform the self-test by using the test pushbuttons on a DECtalk
O,

module.

The self-test control sequence is as follows.

|
S

ESC

;

027

091

053

059

***

121

The first parameter is 5 to specify the DECtalk private test set
The P2 parameter specifies the test to be performed (Table 5-3).

The TEST_POWER parameter (P2 = 1) causes DECtalk to rerun its power-up
initialization and test sequences. All DECtalk operating features return to the

power-up state; the telephone is hung up, the user dictionary is deleted, and all

features are reset to their power-up values.
ARG,

The built-in (canned) message gives you a quick check of the DECtalk system.

The message includes the version number of the DECtalk firmware.
EC

If you try to run an unknown test, you will get a *‘bad private’’ error.
s,

52
[

MAINTENANCE AND TEST COMMANDS

s
S

i
L

e
i

R
.

SlSl

v 0 o

Gli

s
_

i
a0

.
S

. .

) e

‘

e
T

me
o

»3“‘;»

)

mah
e
.

Gl
%?2‘“
L

"
e

o
-

@‘f’%;T
.
e
o

.
s

i t\"'«%
i

5.3.3

;

L
o

.
S

:@f’& g'{;,'m“

‘

.
:

.
i

‘ugg{fi«fl% '

Kol

Device Status Request (DSR) (B

The brief DSR control sequence is as follows.
ESC

027

[

091

053

110

If no malfunctions are detected, DECtalk replies with the following sequence.

ESC

[

027

091

048

110

If a malfunction is detected, DECtalk replies with the following sequence.
ESC

027

[

091

‘

051

Applications can use this brief DSR format in most cases, because a brief
request does not reset any of DECtalk’s internal error flags. The following
extended DSR format is useful when a malfunction is detected.

CHAPTER S5

5.3.4

Device Status Request (DSR) (Extended Report)

AR,

The extended DSR control sequence is as follows.

ESC

[

027

091

110

If no malfunctions are detected, DECtalk replies with one of two sequences. If

this is the first extended DSR since DECtalk was powered up (a ‘“‘rerun”
power-up self-test counts as an initial power-up), DECtalk replies with the
following sequences.

ESC

[

027

091

048

110

ESC

[

027

091

063

050

049

110

For later requests, DECtalk replies with the following sequences.

ESC

027

091

048

110

ESC

[

027

091

063

050

048

110

If a malfunction is detected, DECtalk sends the following sequences.

ESC

[

027

091

051

110

ESC

[

;

027

091

063

059

110

Table 5-4 lists Pn error codes.

NOTE:

The extended status request sequence resets the error flags.

MAINTENANCE AND TEST COMMANDS

wg;% '

. W‘tgy*%% .

.
|i
e
,,

M*W .
e

T %’fi%fi: o N&%’

mfi
¢

G
/
Lt
»;4'@2%
L
.

io o e
e

)

S
&'%}T«

o L g;f; L

.
o

L
.

e
=
=

.
=

o
i

vl %{

e—

-e

.
o

\%@ .

%‘5}&3&&?@,
S

kf%

A
.

L
&

e
o

“

o V%,;fl

.
el
e

5.3.5

Reset to Initial State (RIS)

Tables 5-1 and 5-2 show how RIS affects DECtalk. The RIS escape sequence
is as follows.

ESC

027

099

This sequence resets DECtalk to its power-up state. The character set selections are returned to their default values and the conformance level is reset to
level 2.

All pending, unspoken text is lost. All user-defined dictionary entries are deleted. The telephone is returned to the on-hook state. The RIS sequence always
turns on host speech.

CHAPTER 5

9.3.6

Soft Terminal Reset (DECSTR)

Table 5-2 shows how the soft terminal reset affects DECtalk. The DECSTR
control sequence is as follows.

ESC

[

027

091

033

112

This sequence resets DECtalk to its power-up state. The user-load
ary and the conformance level are not affected.

able diction-

’
AR,

However, as with RIS, the character set selections are returned to

their default
status, all pending, unspoken text is lost, and the telephone returns to
the on-

hook state.

The DECSTR sequence always turns on host speech.

5.3.7

NVR Control (DECNVR)

DECtalk does not contain an NVR (nonvolatile RAM). However, it ‘“‘under-

stands” the NVR control sequence, so that the programs that
attempt to
manipulate the NVR will get the appropriate error responses. The DECNVR
sequence is as follows.

ESC

[

027

091

;P2

059

***

033

114

The P1 parameter selects the type of operation (0 = recall data from
= store data in memory).

The P2 parameter specifies which NVR memory to use (0
memory, 1 = read-only factory-default memory).

memory, 1

= read/write user

Most parameter combinations in the control sequence fail. The only

bination is the “‘recall data from default memory”” when P1 = 0 and

valid com-

P2
= 1. The

function of this combination is identical to the soft reset (DECSTR)

DECtalk remembers the success or failure status of the last NVR
operation

command. A device status request (DSR) sequence can check

this status.

TELEPHONE COMMUNICATIONS \J)
You can connect DECtalk to the public telephone system to provide a dial-up
link between remote users on telephones and a computer application program.

DECtalk sends and receives information as a link between a remote user and

the host computer.
DECtalk communicates with the phone through the voice circuits, passing

spoken data to the listener. DECtalk passes information back to the host

computer as ordinary ASCII characters, and as escape sequences. The user
can communicate with the host computer (through DECtalk) by using the

Touch-Tone keypad, if available.

In the United States and Canada, DECtalk has a built-in telephone line interface

that supports tone detection, pulse and tone (Touch-Tone) dialing, “wink’’

detection, and a simple call progress detection system. Wink detection lets the
host computer determine that a caller has hung up much earlier than is possi-

ble by using only timeouts. The wink detection function permits DECtalk units
to service a much larger volume of telephone data traffic.

CHAPTER 6

6.1

TELEPHONE MANAGEMENT (DT_PHONE)

This multifunction control sequence takes one or more parameters and con-

trols the attached telephone and Touch-Tone keypad

interface. The

DT_PHONE escape sequence is as follows.

ESC

027

;

080 048 059 054 048 059 ***

ESC\

122 027

092

The parameters, starting at P3, act as a list of telephone commands and are

executed in sequence (from left to right). Table 6-1

lists the valid P3

parameters.

The dialing commands have a dialer control string between the z and the string

terminator.
EL

A single DT_PHONE sequence can perform several commands. Some commands (such as enable autoanswer) can take additional parameters.
All DT_PHONE commands return a status report to the host computer in the
following format. Table 6-2 lists the valid P3 status values.

ESC

;

ESC

027

080

048

059

055

048

059

122

027

092

Telephone keypad characters are sent as text, not escape sequences.
The R3_PH_TIMEOUT and R3_PH_WINK reply sequences are sent asynchronously. That is, the reply sequence may arrive as unrequested input, and
A

the application program must be ready to receive them.

TELEPHONE COMMUNICATIONS

s o

g
L

o -

b
o

S
.
L
.
- .

e .
B
.
-

L
i

. gmx%’%

G
e

e
e

.
.
.

i
e

38 .? A

i
)

Lok

\:fl"&fl\;@,’%&é‘i@
o
i

‘

flfi%%s,

%‘«

. ;‘5{

7 )‘33

5: i

L
i
e

)

Shalenby

S
'

iev
&

et
i

. 3{

it '«:.g:w

—

‘:2$
-

oy
e

il
i
.
i
.
e

S
=

iy &;\MX%@

.
o

‘é‘

e=
s

CHAPTER 6
O

ARG

s
S

e
o

o
.

[

~
=

'v;%f(

o
-

i
o

.;:,«

%: 0

SRR

9 %&
.
e

& i

R
e

—

&
G

«g‘&?@%fi:

. ’M@«”“’M“’?‘:
- .

S
e

e
G

e
RS
o
v.«v

. ';w

}é‘a

o
o

.
i

e¥

.
e

% v M’*‘»fé«

N'»‘ 4
e Vfi
e

e
S

e m&;fi%&‘ .
}\Zf%%%‘%%q{.
s

e
0

-=e '?§

.
e

=%%;;i:&%« =i

=
o

= =

. e

S&é

«,

mE:

=
=Sas

=
-

o e"» s

-o

=—

. -

b
i
iz

s
o

‘

/“«‘e’x‘
G

e
e

DR
A

i
3&(“@‘.

-o

=
TR

s
=

/«.«%flw{

-~.o >

Shen

o
Sues

e
o

%?A =

GRS

=
S

e .

=
i

A
bAéi,,
‘M

}‘)

.),
M:tp',e

.
T

SR
o

S K

:<“‘§? =

i
o

i
i
ey

- -

AN,

TELEPHONE COMMUNICATIONS

6.1.1

PH_ANSWER

DECtalk is set up to answer incoming phone calls. The parameter that follows
the PH_ANSWER command parameter indicates the number of rings to wait
before answering the telephone. A parameter of 0 or 1 means answer the
telephone after the first ring; 2 means answer after two rings, and so on.
If the telephone is off-hook when the host sendsa PH_ANSWER parameter,

DECtalk hangs up the telephone (disconnects any active call) before executing

the PH_ANSWER command. This command also disables the telephone
keypad, the timeout, and the wink detection feature.

DECtalk sends two status replies to a PH_ANSWER request. The first status
reply (which is sent immediately) informs the host computer that the
DT_PHONE command was correctly received. The second reply (which is sent
when the telephone is actually answered) informs the host computer that the

telephone has actually been answered.

DECtalk stops waiting for incoming calls whenever the host computer sends
PH_HANGUP, PH_TONE_DIAL, PH_PULSE_DIAL, RIS, or DECSTR.
6.1.2

PH_HANGUP

This command hangs up the telephone. The status reply is delayed until the
telephone is actually back on-hook (disconnected). The host computer should
wait for the R3_PH_ONHOOK reply before sending other commands to
DECtalk.

This command disables the telephone keypad, the timeout, and the wink detection feature.

CHAPTER 6

6.1.3

PH_KEYPAD

This command enables the telephone keypad. The request is ignored if the
phone is inactive (on-hook); however, DECtalk returns an R3_PH_ONHOOK
status reply. The command also disables the timeout function.

The characters received from the keypad look like ordinary ASCII characters.
In a typical keypad, the characters are 0123456789*#. Some keypads have
four extra buttons that generate the characters ABCD. These buttons are

located as follows.

A to the right of 3

B to the right of 6

C to the right of 9
D to the right of #

6.1.4

PH_NOKEYPAD

This command disables the telephone keypad, but maintains the phone con-

nection. This request is ignored if the phone is inactive (on-hook); however,
DECtalk returns an R3_PH_ONHOOK status reply. The command also dis-

ables the timeout function.

6.1.5

PH_STOPPAD

This command enables the telephone keypad, just like the PH_KEYPAD com-

mand, except that it also enables a special autostop mode. While in this mode,

DECtalk simulates ‘‘stop speaking’ and ‘‘disable speaking”’ commands before

the key code (character) is sent to the host computer. DECtalk stops speaking
as soon as possible; the pending speech is discarded.

Control sequences buffered in the communication channel between the host

computer (user program) and DECtalk are processed, but no text will be Spob
z

W %

ken until DECtalk receives the “‘enable speaking’’ command. If a program uses

phonemic mode, extra care should be taken to recover from a ‘‘disable speak=ing” command during phonemic text transmission.
*

»&

Y‘oucan dlsable the a’i’Utostop mode without disabling the keypad by sending a
PH_KEYPAD sequence.

NOTE:

The PH_STOPPAD command is ignored if DECtalk is operating as a

level 1 device. (See Chapter 5.)

TELEPHONE COMMUNICATIONS

This commahd starts (or restarts) an internal DECtalk timer. If the user does
not press a telephone keypad button within the timeout interval, an
R3_PH_TIMEOUT status is returned (Table 6-2).

The applicaiion program should set PH_KEYPAD on before sending a

PH_TIMEOUT command; otherwise, the user cannot respond to DECtalk

requests for input.

The parameter (P4) following PH_TIMEOUT is the timeout value, that is, the
number of seconds to wait for a response from the caller. A parameter of 0
cancels any active timeouts. After a timeout, the timer is stopped. The application program must send a new PH_TIMEOUT command to restart the timer.
Timeouts are the only positive way to detect that the caller has hung up the
telephone, although the wink detector can determine this earlier in many cases.
6.1.7

PH_TONE_DIAL and PH_PULSE_DIAL

DECtalk can dial an outgoing call by using these two commands. If DECtalk is
connected to a Touch-Tone public telephone network, then use the
PH_TONE_DIAL parameter; otherwise, use PH_PULSE_DIAL.
PH_PULSE_DIAL works like a rotary phone dial.
If the telephone is on-hook when the host computer sends a dialing command,
DECtalk picks up the telephone. If the first character of the dialing string is not
a W, then the host computer inserts a standard 2-second delay.

The text between the command terminator z and the ESC \ sequence is the
number to dial. For the Touch-Tone dialing system, the characters
0123456789*#ABCD!AWXTP are recognized. For the interrupted pulse dialing
system, the characters 0123456789!AWXTP are recognized. (Also see Table
A-4.)

The character ! inserts a 1-second delay into the dialing stream. DECtalk
pauses during the dialing sequence every time it finds a ! character.

On some telephone systems, a user can press the switch hook to transfer calls
or otherwise interrupt a phone call. This signal is called a switch-hook flash.
The A character inserts a 0.25-second switch-hook flash signal into the dialing
stream. You can use successive A characters to generate longer flashes.
63

CHAPTER 6
g

With Touch-Tone dialing, the characters ABCD generate the extra four tones of
the military handset. A is the character to the right of the 3; B is the character

below it, and so on.
The character T switches DECtalk to tone dialing mode; while the character P

switches DECtalk to interrupted pulse dialing mode.
The character W is used to wait for dial tone. You can place a maximum waiting
period value, in seconds, after the W by enclosing the number in < > characters. The W range is from 10 to 120 seconds. The following conditions apply.
1.

If there is no < character after the W, then a 30-second default period is

AR,

used.
ARG

If the user-specified time is less than 10 seconds, the W is set at 10

seconds.

3. If the user-specified time is greater than 120 seconds, the W is set at 120
seconds.
4.

If a dial tone is detected during this waiting period, DECtalk proceeds with

dialing. If a dial tone is not detected, the dialing sequence fails with a ‘‘bad
dial tone’ status.
The character X performs call progress detection. The maximum waiting time

for call progress detection can be specified by a value placed between the < >

characters (as in the W command).
The X default period is 30 seconds; the range is from 10 to 120 seconds.

TELEPHONE COMMUNICATIONS

6.1.8

PH_WINK

This command enables the wink detection feature. If a wink (a line current
interruption) is detected on the line, a R3_PH_WINK status report is sent to the

host computer. The command is ignored if the telephone is inactive (on-hook).

On most telephone exchanges, one or more winks will be generated when a
caller hangs up on DECtalk. Winks can be used to clear a call. A timeout must
also be used, to deal with situations where the wink is not available. The
command is ignored if the DECtalk operates as a level 1 device. (See Chapter
5.)

The wink detector may see a false wink, caused by transients on the telephone
line, if the wink detector is enabled immediately after the telephone is taken off-

hook. The wink detector should be enabled a few seconds after pick-up. The

easiest way to accomplish this is to enable the wink detector after the inital
greeting message is spoken.

6.1.9 PH_NOWINK
This command disables the wink detection feature. It is ignored if the telephone

is inactive (on-hook). The command is ignored if DECtalk operates as a level 1
device. (See Chapter 5.)

6.1.10 Examples of Telephone Communication Sessions
The following figures show typical communication sessions between a host
computer and a user.

Figure 6-1 shows a phone call session, including a timeout sequence initiated
because a user was disconnected.

Figure 6-2 is a successful dialout session with a user being connected to the
host computer.

Figure 6-3 is a busyout session where the telephone is busy while the host
computer is trying to call.

Figure 6-4 shows a dialout session where the telephone is not answered.

CHAPTER 6

ESCP0O:60:;10;1zESC\

t\""\k \'{
$

ZnpaW

ESCPO:70:0zESC\
HOST COMPUTER ENABLES COMMUNICATION WITH DT__PHONE/PH__ANSWER.

DECtalk SENDS R3_PH__ONHOOK TO HOST COMPUTER.

ESCPO;70:1zESC\

HEwoO.

DECtalk...

ESCPO0;60;30;20zESC\
A

CALLER DIALS IN. DECtalk SENDS R3_PH_OFFHOOK TO HOST COMPUTER.
HOST COMPUTER ENABLES 20-SECOND TIMEOUT AND BEGINS COMMUNICATION CONTROL.

ESCPO;70;:2zESC\

CASE 1: USER HANGS UP. AFTER NO SIGNAL IS RECEIVED FOR 20 SECONDS,
DECtalk SENDS R3_PH__TIMEOUT TO HOST COMPUTER.

ESCPO;60;11zESC\

ESCPO;70;0zESC\

CASE 2: HOST COMPUTER TERMINATES SESSION AND SENDS PH_HANGUP.
DECtalk HANGS UP PHONE AND REPLIES WITH R3_PH__ONHOOK.
MA-0350-85
A

Figure 6-1

Phone Call Session with User Disconnect

TELEPHONE COMMUNICATIONS

ESCPO:60:402W.5551212
X ESC\

HOST COMPUTER SENDS DIAL SEQUENCE.
DECtalk WAITS FOR DIAL TONE.

DECtalk DIALS NUMBER.

ESC\
ESCPO:60:402zWH5551212X

DECtalks WAITS DURING RING BACK.

ESCPO:70;1 zESCN
TELEPHONE IS ANSWERED.

DECtalk SENDS R3_RH_OFFHOOK TO HOST COMPUTER.

Figure 6-2

MA-1164-8B5

Successful Dialout Session

CHAPTER 6
P

ESCPO0:60:40zWi5551212
X ESC\

HOST COMPUTER SENDS DIAL SEQUENCE.
DECtalk WAITS FOR DIAL TONE.

ESCP0O;60;402zW5551212:X ESC\

ESCPO;70;60 zESC)\

DECtalk HEARS BUSY SIGNAL AND
SENDS R3_RH_BUSY TO HOST COMPUTER.

Figure 6-3

Ma-1163-85

Busyout Session

TELEPHONE COMMUNICATIONS

m\"“%

HOST COMPUTER SENDS DIAL SEQUENCE.
DECtalk WAITS FOR A DIAL TONE.

&
S

N
by

,“a'fiw”‘“

Zia®

‘l‘ i

DECtalk HEARS RING BACK.

LISTENS TO IT FOR 30 SECONDS.

z ESC\
ESCPO:70:7
DECtalk GIVES UP AND SENDS

R3_RM__NOANSWER TO HOST COMPUTER.

Figure 6-4

MA-1165-85

No Answer Session

-%4\=wm.mu%mk.Smf.w,mo.:bmo%Tm.,e(?&;7M)w,wR-B=.fimxeo2B,x»w»?2

Nfiqfiamf»,.fow\.u%.%,ny&.?,.:.W-Q%W-w-@z.cwwm,;.\;M.{«fm%inwxw.,nwm%fi,.m=m«,M.=fi,‘,Mm.aw.wmw/w.&mMu,,uw4mxbfiwsJnw;w»/SwcEwflnwwv%w.ai:&,M&Wf%W«):-o,?efma,-.s3S

MNf.,}wiéwnfr=Wl,&-m@wé«ffi%if-9%%2Nw&‘f»i:w..fu\un“yw\:{a.zf»wf-,=.fw=&l,-w.},%iu.m«ww‘,&4w.?..Smaf%.eSm.:k..=pqBe=QVffil..ffix¢wti(\mf;fio,aM-W«L.- NJ)Avw%,.w“wm%.snw«azéf=ina.Swe@fSi=av\,w}A.D«ew.S%hSe{amsmwuwj«‘s4S/=@Hw.“.iw:@%.A«(o..)mS.w,?.@=“,.w.o%,hfSu«w,emwt»fi?foi.&we:%-t=»wm%)finwq«.u%s.:-..,uc=m5wu,n\S

3fiwv.@ S 4=%:?mo = =w%wowm i- =a - o=

-r.»mwv e s,.v%.uki Ss - 7 Bi = o wm,w e oA

.!f-Mi;l»

6.2

a carriage return to th e en

CHAPTER
6
P

T c ® o T3 < 0p) X O Q £ & © - © oQ c js o 2} < o 3 ()] L O ot © = n Q c ° n @ o\ O ] ® w @ - O - o@ o © c ©
Q
o
keyp adc haracters to the host computer This com man d allows DE Ct a e e oa d

KEYPAD MASK (DT_MASK)

of ts repl ies

The comman d t akes o n e par a meter

w h IC h

o .:-W

«,

4
i

wS,esRin.‘.tuym

’“fi ‘»,%1L
o

Pzz"‘u
:a

is in terpreted as a v alue from the

. S,

“g,,x i

f:‘:\v
et

'i"’i.

%w -

-=2rebSw.f,mu:.‘w.Rr:fa.mxiv?um}.‘,

. W»z,; »s

Amf‘i.x!

16-bi keypad mask matr IX (Table 6 3)
i

p—

TELEPHONE COMMUNICATIONS

If a bit
is set, DECtalk sends a carriage return after sending the associated

keypad character. If any bit is set, DECtalk sends a carriage return after its

escape seguence replies. (The carriage return follows the ESC \ string terminator.) Many operating systems translate the carriage return to a carriage
return and line feed. The DT_MASK escape sequence is as follows.

ESCP

027 080 048 059 056 051

059 ***

;

059 **

ESC
\

122 027 092

The P3 parameter is bit-encoded. Specified values have associated characters

(Table 6-3). If you specify a value, DECtalk sends a carriage return after the
associated character (when a user presses that key).

The P4 parameter is the level 2 device function. It controls the interpretation of
the P3 parameter as follows.

If P4 is 0, then P3 is a new keypad mask.
If P4is 1, then P3 is a bit set to be set in the keypad mask.
If P4is 2, then P3 is a bit set to be cleared in the keypad mask.

For example, to have DECtalk treat the # and * characters as response terminators (but not the digits), a program would send the following sequence.

ESCP ;
8
3
:
3
027 080 059 056 057 059 051

ESC
\

048 055 050 122 027 092

(3072 = 1024 + 2048)

If the person calling the application presses 123# followed by a keypad timeout, DECtalk would send the following.
1 2 3 # <carriage return>

ESC P

;

ESC

027 080

059

055

048

059

050

122

027

092

CHAPTER 6

This allows the application program to use standard line-oriented input routines, rather than character-oriented routines. If you specify a P3 parameter =0
with DT_MASK, DECtalk will not send a carriage return after keypad characters or escape sequences.

NOTE:

DECtalk will send a carriage return after all sequences, including

responses to non-DECtalk-specific sequences such as device status request.
At power-up, the keypad mask is set to all Os or all 1s, depending on the

position of DIP switch SW3 on the DECtalk module. (See Appendix B.)
DT_MASK simplifies application development when DECtalk is connected to a
host computer via a packet-switched network or a network using the SNA

(systems network architecture) protocol. These networks have a significant

overhead associated with each message, so sending a line of text (several
characters) is more economical than sending a single character.

DT_MASK is also useful when DECtalk is connected to an operating system
that prefers to communicate line-by-line, rather than character-by-character.
For example, when DT_MASK is on, you can use BASIC’s INPUT LINE command to read text from DECtalk.

DECtalk PHONEMIC INPUT

This chapter describes the phonemics (English sound system) used in DECtalk

and the ways to control DECtalk’s pronunciation.
7.1

PHONEMIC TRANSCRIPTION

Most users do not need to know anything about DECtalk phonemics input.

However, to understand how the DECtalk system works and to make DECtalk
correctly pronounce any English word, you need to know something about

speech sounds and how to represent them on a keyboard. Because spelling in

English does not always show exactly how words are pronounced, dictionaries
use symbols to show how words really sound. Sometimes these symbols are
the same as letters used in spelling. A word written the way it is pronounced is

said to be written in phonemic transcription.

7.2

PRONUNCIATION ERRORS

When DECtalk says a word or phrase incorrectly, you may need to use phonemic input to get the desired pronunciation. The following list suggests the most
common types of errors that DECtalk makes, and the best corrective action.

1. DECtalk mispronounces a proper name.
Lee lacocca

Corrective action: Convert to phonemic form.
Lee [ayaxk'owkax]

Or misspell in a clever way.

DECtalk PHONEMIC INPUT

2. DECtalk mispronounces an acronym.
The UN building

Corrective action: Respell with spaces between the letters.

UN '
Or use phonemics.
['yu 'ehn]

3. DECtalk mispronounces an unfamiliar word.
articulatory

Corrective action: Convert to phonemic form.
[aart'ihkyxaxlax owriy]

4. DECtalk mishandles'a letter string containing nonalphabetic characters.
sys$system
vax750

Corrective action: Respell with inserted spaces.
sys $§ system
vax 7 50

Or convert to phonemic form.

[s'ihs d’aalrrs ayn s’ihstaxm]
[v’aeks s’ehven f’inftiy]

CHAPTER 7

5. DECtalk guesses wrong for an ambiguously prdnou
nced

word.

The insert

Get the lead out.

Corrective action: Request alternate pronunciation

notation.

by using parenthesis

The )insert

Get the )lead out.
Or convert to phonemic form.
The ['ihnsrrt]

[g’eht dhax I'ehd "awt]
DECtalk uses the wrong syntactic classification

of a preposition/particle.

He takes on tough jobs.

(“He does tough jobs” versus “He accepts graft

when on tough jobs.”’)

Corrective action: Add a stress phoneme when needed

He takes ['Jon tough jobs.

Or convert to phonemic form.
He takes ['aan] tough jobs.
DECtalk uses the wrong phrasing.
Following a long gasp shouts were heard.

Corrective action: Add commas or verb-phrase
introducer phonemes

where needed.

'Followinga long gasp, shouts were heard.

DECtalk PHONEMIC INPUT

7.3

INTRODUCTION TO PHONEMIC THEORY

At one timé long ago, English was pronounced as it was spelled, with each
letter (or pair of letters) representing one sound, or phoneme. Because of
historical sound changes (such as the dropping of sounds like the “gh” of
“bought”’) and word borrowing from other languages, English pronunciation
rules have become complex and include many exceptions.

For example, “of’ is pronounced
with a v sound, while all other English words
spelled with “‘f”’ are pronounced with an f sound. The vowel digraph “ea’” can
in the
be pronounced in at least a half-dozen ways, as illustrated by the sounds

words ‘‘cheap,” ‘‘head,” “‘earth,” and ‘‘idea.”” The letters “th” can be pronounced with a voiceless phoneme, as in “‘thin’’; or with a voiced phoneme, as
in “‘the’’; or the “‘th’’ can represent the ¢ phoneme followed by the A phoneme in
compound words such as ‘“‘pothole.”

Some words have two pronunciations, for example, “‘read.”” Correct pronuncia-

tion of a sentence such as “Will you read the book or have you read it
already?’’ requires an understanding of the meaning of the sentence - a task
beyond DECtalk’s ability. Therefore, DECtalk speaks words with alternate
pronunciations by choosing the more frequently used pronunciation.
You can get the alternate pronunciation in several ways.

e By misspelling
the word: “red” for “‘read”
e By phonemic spelling: [r'ehd]

e By using the alternative pronunciation symbol [)], as described in Section

7.6.10. For example, ‘“He )read this book yesterday."

Stress is am important part of phonemic representation. Stress alone distinguishes the two different pronunciations of words like “insert.”
English words usually have one syllable that is spoken with more stress than

the other syllables in the word. You can indicate this primary stress to DECtalk
by placingifi the phonemic symbol ['] before the vowel. The ['] symbol is
described in Section 7.6.1.

CHAPTER 7

For example, the word “‘insert”” can be spoken as a noun

“insert’”’ = ['ihnsrrt]
and as a verb

“insert” = [)]insert = [ihns’rrt].

Considering the complexity of English pronunciation rules and the number of
exceptions, it is not surprising that DECtalk sometimes makes pronunciation

errors. You can adjust DECtalk’s pronunciation through a large number of
symbols, described in the rest of this chapter.
7.4

PHONEMES

A phoneme is the smallest unit of speech that distinguishes one word from

another. Of all the sounds that human beings can produce, relatively few are

significant in any one language. Only about 40 different sound types (phonemes) are used in General American English.

The phonemes of English are not pronounced the same by every speaker. We

all know people who pronounce some words differently from the way we do,
yet we understand them. The differences may occur because we come from

different parts of the country. Because of these variations, there is no such
thing as a universal standard pronunciation of English. DECtalk attempts to
mimic a midwestern (Northern Milwaukee) dialect.

Because DECtalk pronounces a phoneme in a standard rule-governed way, it
is not possible to imitate all other English dialects (although you can approximate some dialectal differences by phonemic spelling).

The following sections describe the vowel and consonant phonemes, stress
and syntactic symbols, and optional direct control of intonation or singing.

DECtalk PHONEMIC INPUT

7.5 VOWEL AND CONSONANT PHONEMES
Linguists have identified about 17 vowel phonemes and 24 consonant pho-

nemes for American English. Figure 7-1 shows the vowel phonemes and the
positions of the tongue when making the vowel sounds. Tongue position (high
versus low in the mouth, and front versus back of the mouth) correlates with
the frequencies of the two lowest natural resonances of the vocal tract. The
lowest resonance frequency, or first formant, issymbolizedby F1 in the figure,
and the second formant by F2. Consonant phonemes are shown at the bottom
of Figure 7-1 with their places of major articulatory constrictions and the manners of forming the constrictions.

FRONT (F2 HIGH) ~=

HIGH

VOWELS
—

» BACK (F2 LOW)

(F1 LOW)

| oUW
IH

MID

LOW

AO
AE

(F1 HIGH)

PLUS THE DIPHTHONGS:

PLUS THE SYLLABIC CONSONANTS:

CONSONANTS
LABIAL

STOPS

DENTAL

NASALS

]
e

FRICATIVES

ALVEOLAR

PALATAL

VELAR

8
|

‘

GLOTTAL

'
.

AFFRICATES

GLIDES

LIQUIDS

R
MA-7627-83

Figure 7-1

Phonemes Classified by Their Position in the Mouth Cavity

CHAPTER 7

Table 7-1 lists the vowel phonemes and Table 7-2 lists the consonant phonemes. The DECtalk symbols used for each phoneme are identified by a key
word with the relevant phonemic sound in italic.
iy

“;‘

e
&

e
s

s
5

e
1y

o
:

‘

(RS

Q,@

o
.
|

o
.

n
‘
e

‘

G ab*/ '«’g‘%m )

e y,,'«@f

e
e

e
v@%;%fifl
N
-

e ?de (W w

e \"rl
e
=
e

S v'é,“;&w
5

ot
o

e
)

S
)

w%‘

L
R AR
A
AR
A
5
St
o

L
e

‘o

.
s
=
e

NERR

?,"‘:

;":‘,(’.
i

’

i
G

L
i

Lo
e

§e

i
G

%‘&W
e

-7

L %f‘l‘& B

[

;

i
(S

;

%{?

: w@

G
s

kicepe

;

.
o

L
i

)

- i

o
:

)

w
i

;

o
«2";,;}“:
Pt

‘

g?m o

‘Ws%‘#'

Pt e
i

3%%

S
s

o
b

“-v!fs‘.:

.
.

‘w,

'g'fi o e

-n.{?\za\(t,«giémflg%m,

;

?’@w

4:.

i
i

"‘ég e

.
o
Cn

i %’«“

o
b

w‘{%‘*

%@;J

‘

%fi;«r

Q:%q
éé‘%
e

o fifl

o
i

‘«,}v

ety

J—

DECtalk PHONEMIC INPUT

In many cases, phonemes are indicated by two letters, instead of special
characters or diacritic symbols that often appear in dictionaries. DECtalk
requires a case-insensitive representation (uppercase and lowercase are
acceptable). The letter pairs have been designed so that it is not necessary to

put a space between phonemes of a word. In fact, the space indicates word
boundaries. DECtalk can parse input phonemic letter sequences to determine
the unique phoneme sequence in all cases.

Phonemes are enclosed in square brackets instead of between the more traditional / symbol. The [ and ] characters mark the beginning and end of phonemic
mode clearly with distinctively different symbols. The input format is not strictly

phonemic because it also permits you
to enter certain allophones (variants of a
phoneme), making the representation closer to a broad phonetic transcription.

vyu" e
g

0
'

o
G

. .i

.
e

.
R

il i

v,;v?
.

‘is?%'%"‘ .

\,%@?@fi

.
.
i

{3‘««««7%;«. W%
.

. -

.
S

S
o

.
.
v,w‘;’@'/

'Vw%‘

g‘

%}'

i
.

“

“;%sg?v&w

e
e

.
mfivx

%@% e

...

;

L
.

G
cE

o
.

‘

o ”w‘é‘f:

...

CHAPTER 7

7.5.1

Phonemic Correction the Easy Way

Most users will want to learn the phonemic code. However, you can also
consult one of the dictionaries described in Appendix E to determine the pho-

nemic pronunciation for a word that DECtalk gets wrong. You can convert the
word to the DECtalk phonemic code by using the table in Appendix E.

For example, according to the Merriam-Webster Dictionary, the pronunciation

of the word “Mozart” is

\'mot-,sart\

Consulting Table E-1 of Appendix E, you can convert this transcription to the
DECtalk phonemic string
[m’owtsaart]

7.5.2

Automatic Phonemic Substitution for a

Mispronounced Word
Every time DECtalk mispronounces a word in running text, your host computer

could be programmed to replace the text string (‘‘Mozart’’) with a phonemic
string ([m’owtsaart]).

However, if the number of words requiring phonemic translation in an applica-

tion is small (under about 200), it might be simpler to download a user dictionary to DECtalk and let DECtalk perform the replacement automatically. A sample user dictionary is shown below.

Letters

Phonemes

Mozart

m’owtsaart

logo

'aog awt

You can download the dictionary to DECtalk with the following commands.

ESC

;

Mozart m’'owtsaart ESC \

ESCP

;

logo I'aog awt ESC
\

DECtalk PHONEMIC INPUT

NOTE: Keep your user dictionary as a file on the host computer because the
dictionary must be downloaded every time you power up DECtalk. For more
information on user dictionary creation and downloading procedures, see Section 8.9.

The white spaces between the components of this command are there for
clarity. Do not type spaces in your actual command, except at the position
indicating the separation between the word and its phonemic representation.
Once downloaded, the dictionary overrides the standard DECtalk dictionary or
letter-to-sound rules that would otherwise handle these words.

DECtalk will speak the plural form of a word in the user dictionary by stripping
off the “‘s” and consulting the dictionary for a match to the remaining root.
Similarly, DECtalk will remove many common suffixes such as “-ing” and
“-ed” by rul@ and compare the root with entries in the user dictionary. With this
function, you do not need to place all of tha derived forms of words in the user
dictionary.

You should be aware of some pitfalls when using the dictionary. Placing a word
in the dictionary could cause a similar word to be mispronounced. For example,
if you place ‘‘on” in the dictionary, DECtalk might decompose a word such as
“only” as ‘““‘on” plus “-ly,” and mispronounce the word. The only way to correct
this problem is to add ‘“‘only’”’ to the user dictionary. It is difficult to foresee
these cases, but they are relatively rare.
7.5.3

Vowels

The vowel phonemes of English are listed in Table 7-1. While DECtalk recognizes 17 vowel phonemes, these vowels can sometimes change slightly when
surrounded by certain phonemes. These variants are discussed in Section
7.54.

7.5.4

Vowel Allophones

The variants of a phoneme are called allophones. For example, a vowel such
as the [iy] in ‘‘beat” sounds slightly different when followed by [r] as in ““beer.”
Sections 7.5.4.1 through 7.5.4.3 describe some allophones.

CHAPTER 7

7.5.4.1

Allophones for R-Colored Vowels - The vowels in words such as
“beer,” ‘“bear,” “bar,” “bore,” and “poor’ are different from the available

vowel phonemes in DECtalk. They require special vowel-r allophones, which

are listed in Table 7-1.

7.5.4.2

Schwa Allophone - Another problem is with the unstressed reduced
vowel called ““schwa’’ in English. The vowel appears in words such as “about’’

and “kisses.” In “‘kisses,”’ the vowel is produced with a higher tongue position,
symbolized by the vowel allophone [ix] in Table 7-1. You can choose between

[ax] and [ix] by noting the characteristics of the adjacent phonemes, but listening to the words will result in the best choice.

7.5.4.3

Syllabic Consonants - The final syllable in words such as “‘butter,”
“bottle,” and “‘button’ is usually symbolized in a dictionary as consisting of a

short vowel followed by a consonant. For better sounding synthesis, DECtalk
uses a set of syllabic consonants, [rr], [el], and [en] of Table 7-1, that are
realized without the short schwa. Syllabic ‘‘r’’ shares the same symbol as the

phoneme [rr] in a word such as “‘bird,” but this leads to no confusion inside
DECtalk.

The [em] allophone used in an earlier version of DECtalk no longer exists and
must be replaced by the two-phoneme sequence [axm] as in the word “‘bot-

tom’ = [b’aataxm].

In most situations, you do not need to be concerned about allophones because
the vowel phonemes in Table 7-1 will be automatically converted into the
appropriate allophones by DECtalk rules. For the sophisticated user, allophone

selection can be induced or blocked by using the syllable boundary phoneme

[-] and the rule-blocking phoneme [~] (Sections 7.6.7 and 7.5.6.6), or by
inserting allophone symbols in the phonemic spelling.

DECtalk PHONEMIC INPUT

7.5.5

Consonants

The symbols that represent consonants are straightforward (Table 7-2). In two
cases, [hx] and [yx], the two-letter sequence ensures unambiguous parsing
because the letters “‘h’’ and ‘‘'y”’ are parts of some vowel symbols.
DECtalk speaks an English dialect that does not distinguish between voiced
and vonceiefis w, so words like “‘which”’ and “‘witch’’ are pronounced alike as
[w’ihch].

The letter ‘‘g” can be pronounced in two ways. In words like “gift,” the consonant phoneme [g] is used. In words like “‘gin,” the phoneme [jh] is used.
The letter sequence ‘‘th”’ can be pronounced with a voiceless sound [th] as in
“thin”’ or with a voiced sound [dh] as in “this.”

The [y] phoneme used in an earlier version of DECtalk has been replaced by
the [yx] phoneme (as in the word “‘yet”’ = [yx'eht]).
7.5.6

Consonant Allophones

The consonants [t], [d], [r], and [I]] may be replaced by special allophones under
certain conditions. These allophones are listed in Table 7-2.
7.5.6.1 Dental Flap - The [t] and [d] phonemes are often replaced by a very
brief tongue flap allophone [dx] when the consonant phoneme appears

between two vowels and the second vowel is unstressed. DECtalk rules insert

this allophone in appropriate situations.

7.5.6.2 Glottal-T - The m phoneme may be replaced by a glottalized [tx]
allophone, especially in word-final position if the next word begins with a
sonorant consonant. DECtalk rules insert the allophone where appropriate.
7.5.6.3 Postvocalic-R
- The [r] that appears after a vowel is not as constricted as a word-initial [r]. DECtalk automatically selects this somewhat velarized allophone [rx] or an r-colored diphthong where appropriate.

CHAPTER 7

7.5.6.4 Postvocalic-L - The [l] that appears after a vowel may sound
different
from the [I] in other contexts. For some speakers, the tongue tip may not even
reach the roof of the mouth. This postvocalic allophone [Ix] is automatica
lly
selected by DECtalk.

7.5.6.5

Gilottal Stop - The glottal stop [q] is used in some situations to indi-

cate a word boundary, especially when the next word begins with
a vowel.

Overuse of this symbol can lead to a stilted style of speaking.

7.5.6.6 Controlling Allophone Selection - DECtalk automatically
inserts certain other allophones for [k], [q], and [nx] when appropriate. DECtalk
also
selects the prevoiced and voiceless unaspirated allophones of [b], [d], and [g].

You cannot access these allophones.

If DECtalk does not select one of the allophones of Table 7-2, you
can insert
the allophone symbol directly in a phonemic representation of the word
in

question.

If DECtalk uses one of these allophones inappropriately, place the rule-blocking phoneme [~] before the phoneme in question to block application
of all
allophonic substitution rules. For example, to say “‘batter’’ without a flap
being
substituted for the [t], enter the phonemic string [b'ae~trr].

7.5.7

Silence Phoneme [_]

DECtalk automatically inserts a silence (brief pause) whenever punctuatio
n
appears in the text. The phonemic silence symbol [_] is useful for controlling

silence while in phonemic mode. Silences and other pauses are described
more detail in Section 7.7.1.

7.6

STRESS AND SYNTACTIC SYMBOLS

Correct speech is more than simply stringing together a series of words

phonemes. The meaning of a sentence is carried
by the words, plus rhythm,

stress, and intonation (pitch change). You recognize a question by
the rising
intonation of the voice, while a statement is usually accompanied
by falling

intonation. A speaker can give certain words in a sentence more importanc

e by
adding stress (loudness and length) to them. Pitch often reveals the emotional

state of the speaker. For effective communication, you need to consider

expressive features as well as the segmental features of speech.

these

DECtalk PHONEMIC INPUT
As any good actor knows, punctuation alone is not enough to indicate the full

meaning of a sentence. Some fine points of expression cannot be indicated by

using phonemic symbols. Full control of the expression of a sentence is gained

by directly changing the duration and pitch of words and phrases. These direct

control techniques are discussed in Section 7.7.1.

DECtalk uses stress and syntactic symbols to control aspects of rhythm,
stress, and

intonation patterns. These symbols include punctuation marks

such as commas, periods, and exclamation marks. Punctuation marks are
recognized by DECtalk as indicating special phrasing requirements. The following sections and Table 7-3 show how to improve the phrases in DECtalk
speech.

.
'“@,?}mm
..
s
.i

- ‘é‘;fi “4." L

L
-

e %5,« i
el

e
.

L
.
i

Y
.
i

L
o

. :

e
=-

b
e . ‘gg"‘w
i
.
.

i ‘5;'@;3;) i

L
.g

|
i

—-

vesl

%k;

.--AT

.
i

Hee

%.éi-%gf

=
.

;ég;

.
k:;w‘@%
o

o
=

.
e
.

\f:?"}:r%fi:‘

CHAPTER 7

7.6.1

Primary Stress [’]
All content words of English (nouns, verbs, adjectives, and adverbs) contain

one primary stressed syllable. DECtalk represents primary stress on a syllable
with an apostrophe ['] placed before the stressed vowel phoneme of the word

as in the following example.

[bahtrr].
(No stress, flat intonation, too rapid.)

[baht’rr].

(Stress on wrong syllable - “butter.”)

[b’ahtrr].
(Correct stress - “butter.”)
A,

You can also place the primary stress symbol between words, in which case

modifies the next word. For example, in the sentence ‘‘He rang up the sale,”

DECtalk treats “‘up” as a preposition (without stress) instead of a particle.
“Up” is correctly stressed if you write the sentence as
He rang ['Jup the sale.

NOTE:

There can be no space between a stress and syntactic phoneme (for

example, [']) and the following word.

DECtalk PHONEMIC INPUT

7.6.2 Secondary Stress [ ]

Use the secondary stress symbol [ ] to indicate a degree of stress that is
between primary stress and umatmssw Secondary stress is appropriate in the

following ca%s

e ToO mgnight the next stror gaat syllable of polysyllabic words, such as
“‘demonstration.”

[d ehmaxnstr’eyshaxn].

e On second parts of compound nouns, as in “pushover.”
[p'uhsh# owvrrl].
e In some pronouns, such as “I'" and “we.”

DECtalk realizes secondary stress by lengthening the vowel sound more than
unstressed (but less than primary stress). A pitch rise may also occur on an
early secondary stress. In most cases, you can leave out the secondary stress
symbol.

CHAPTER 7

’

7.6.3 Emphatic Stress [’]
You can place the emphatic stress symbol ['] before any vowel

sis to that syllable of the word. Good readers of English text
message of the sentence well enough to pick out the

to give empha-

understand the

most important word and

emphasize it. DECtalk merely pronounces words: it does not
understand the
sentences it is saying. DECtalk cannot place emphasis
on words to give a
completely different meaning to the sentence unless you
use the emphatic
stress symbol. Here is an example.

Dennis loves Mary.

(Usual neutral pronunciation.)
[d”ehnihs] loves Mary or [’] Dennis loves Mary.
(Dennis - not Frank - loves Mary.)

Dennis loves [m”ehriy] or Dennis loves [IMary.
(Dennis loves Mary - not Jill.)
The exclamation point has the similar effect on the final

stress of a sentence.

Help!

DECtalk PHONEMIC INPUT

7.6.4

Unstressed Syllables

The English language contains a set of words that are always unstressed,
called syntactic function words. This set includes the following words.
Prepositions (for, over)

Conjunctions (and, but)
Determiners (the, some)
Auxiliary verbs (is, has)

Pronouns (her, myself)
Clause introducers (which, that)
These words have no stress symbols in their dictionary entries. It is sometimes
necessary to emphasize a function word that is stored in DECtalk’s dictionary

without stréss. You can do this by including a primary stress symbol or an
emphatic stress symbol in the phonemic transcription as in the following
example.

He went ['owvrr] the fence, not under it.
or

He went ['Jover the fence, not under it.

CHAPTER 7

7.6.5

Pitch Control [/], [\], [/\]

DECtalk contains built-in rules to determine the pitch contour of a sentence.
While these rules are correct most of the time, you can override them by
placing the pitch rise [/], pitch fall [\], and pitch rise and fall [/\] symbols before

selected words (or vowels if you want finer control).
The [/] and [\] symbols must alternate, and the first symbol must be a rise.

Note that you can place both a rise and a fall on the same syllable by using [/\].
You can hear the difference by trying the following two sentences.

It's a mad mad mad mad world.
It's a [/lmad [\]Jmad [/]mad [\]mad [/\]world.

7.6.6

Word Boundary

Any whitespace character (space, tab, or carriage return) in the text indicates a
word boundary. DECtalk uses word boundary symbols to select the word-

beginning or word-ending allophone of a phoneme.

Some host computers automatically insert a carriage return into lines that are
too long (and would go off the edge of the screen or paper). This may cause

DECtalk to pronounce text incorrectly if a carriage return occurs in the middle
of a word. You can prevent this problem by breaking long sentences with a
carriage return at an appropriate place.
7.6.7

Syllable Boundary [-]

DECtalk uses a set of rules to determine where words break into syllables, so
consonants within words are assigned to their correct syllable. Use the syllable

boundary symbol [-] to tell DECtalk where to assign the consonants within
ambiguous words. (This type of error rarely happens in DECtalk).
Example: oration

[ow-r'eyshaxn] (DECtalk made an incorrect guess.)
[ow-r’eyshaxn] (Correct.)

DECtalk PHONEMIC INPUT

Morpheme Boundary [*]

7.6.8

English words are made up of meaningful units called morphemes. For example, “‘spell’”’ has only one morpheme, while “misspelling” is made up of three:

“mis,”’ ”spé:ll,” and ‘‘ing.”
¥y

In most cases, the pronunciation of a word does not depend on morpheme
boundaries. There are exceptions, however, in which case the morpheme
boundary symbol [*] can be used to force the correct pronunciation. For example, “misspelling” should be pronounced with a double “s.” Adding the morpheme boufldary symbol improves the pronunciation of the word.
misspelling.

mixsp’ehlixnx (text to phoneme translation by DECtalk).

(The single ““s” is too short.)
[mixs*sp’ehlixnx]

(Better.)

7.6.9

Compound Noun [#]

Compound words, such as “‘rush-hour,” should be spoken with less stress on
the second word. Also, words that were once compounds, such as ‘“‘sideache,”
require decomposition for correct pronunciation.

DECtalk’s dictionary includes an extensive list of compound words. You can
use the compound noun symbol [#] to correct compounds that are not in the
dictionary. For example, for ‘‘sideache,” type the following phonemic
transcription.

[s’ayd# eyk].

Using a hyphen in cmmpOund words, for example, ‘“‘rush-hour traffic,”” produc-

es the correct pronunciation most of the time. You rarely need the [*] and [#]
phoneme symbols.

CHAPTER 7

7.6.10

Beginning of Verb Phrase [)]

Moderately long declarative sentences are usually spoken as if they contain
two units: a noun phrase and a verb phrase. There is no pause between these

two phrases, but there is a slowing down at the boundary, and the pitch tends

to fall and then rise. DECtalk searches for this syntactic boundary to change
pitch. However, the rarity or ambiguity of some verbs can cause confusion.

The old man in the chair was rocking slowly.

(Correct verb phrase detected.)
The old man in the chair sat rocking slowly.
(Verb phrase not detected; pure mechanical analysis of
the sentence does not show where ‘““‘sat” belongs.)
The old man in the chair [)s’aet] rocking slowly.

(Phonemic correction.)

The right parenthesis [)] symbol is useful where a separation is needed
between phrases but a comma is too strong. For example, you can use D] to
indicate a dangling prepositional phrase.

She hit the man with the umbrella.
(The man carries the umbrella.)

She hit the man [)] with the umbrella.
(She uses the umbrelia.)
The [)] symbol has a second function: it can indicate alternate pronunciations of
homographs. Homographs are words that are spelled the same but pronounced differently. For example, the word ““insert’ is either a noun or a verb.

As a noun, it is pronounced ['ihnsrrt] and as a verb it is pronounced [ihns’rrt].
DECtalk typically defaults to the more frequent pronunciation. If DECtalk pronounces ‘‘insert’” as a verb, and you need the other pronunciation, simply place

the ) symbol in front of the word.

The experienced secretary inserts more )inserts per hour.
Appendix D lists the pairs of common homographs that DECtalk knows.

DECtalk PHONEMIC INPUT

Clause Boundary [,]

7.6.11

When a santence is composed of more than one clause, it should be spoken in
such a way gthat the listener can easily separate the sentence into its compoThe comma [,] is the symbol used to indicate clause boundaries.
nent clauses.
An crthagraphic comma and a phonemic comma have identical impact on the
acoustic realization of a sentence.

Insertmg a mmma improves the quahty of spoken sentences in the following
cases.

e After an introductory prepositional phrase:
In particular cars cause po lutmn
(Poor phrasmg )

In pamou@ar cars cause pollution.
(Correct.)
e Around a parenthetical remark:

A picture it seems is worth - - (Poor phrasing.)

A picture, it seems, is worth - - (Correct.)
e In a list
of more than two items:

They ate apples oranges and bananas.
(Poor phrasing.)

They ate apples, oranges and bananas.
(Correct.)

CHAPTER 7
R

7.6.12

Period [.]

A sentence is usually a single, complete thought. It is also the longest utterance
that you can comfortably speak in one breath. DECtalk inserts a pause when it
finds a period that marks the end of the sentence, duplicating the human
speaker’'s pause to take a breath.

The [.] symbol also tells DECtalk that a complete sentence has been sent and it
is safe to begin speaking. DECtalk will not speak a phrase until it finds a

comma or sentence terminator. After a few seconds, text sent to DECtalk is

spoken even if no comma has been received. DECtalk also tests each period to
make sure it is not part of a known abbreviation.

7.6.13

Question Mark [?]

The simplest way to indicate a question in English is by a rising tone at the end

of a sentence, although true question intonation is not that simple and depends
on the meaning of the question.
There are many cases in English where a question (rising) intonation is not
appropriate, even though the sentence ends with a question mark.
e

Rhetorical questions or quotations may contain a question mark,

but the

speaker ends with a period (falling tone).

e Sentences that begin with “wh’ words (‘“who,”’ “what”") usually end with
a falling tone, even if they are questions.
Although DECtalk is smart enough to recognize “‘wh’ questions and speak
them correctly, you should send question marks to DECtalk only when the
sentence really is a question. Here are some examples.

Laura ate her broccoli?
(DECtalk asks a question.)

What time is it?

(DECtalk recognizes “wh” and does not rise at the end).

DECtalk PHONEMIC INPUT

7.6.14

Exclamation Point [!]

Exclamatmms are short statements spoken wwm special emphasus DECtalk
interprets an exclamation point to mean that the last stressed syllable in the
sentence should have extra emphasis.
Stop!

Long sentences ending with an exclamation point typica lly have a single word
that receives extra stress. DECtalk has no way of knowing which word to
stress and chooses the last word by default. Use the emphatic stress symbol
[''] to emphasize a different word when the last word is not appropriate.
Joan won the marathon!
(DECtalk emphasizes the last word.)

['] Joan won the marathon.
(Correct.)

7.6.15

New Paragraph [+]

The new paragraph phoneme [+] should be inserted in text wherever a text
preprocessor thinks that a new paragraph has begun. (DECtalk does not do

this automatically because thereis no standard new paragraph mducator in
general temt - the tabis used m too many other ways )

The new ;jaragraphx phoflame [+] modifies the intonation contour and adds
to running text. The first sentence of a new paragraph is produced with
variety

a higher, more lively fundamental frequency. DECtalk will also pause longer
between paragraphs to give the listener an indication of a change of topic.

[+] This p giragraph has the [+] phoneme inserted in the appropriate place. The
new paragraph symbol can be used in other situations, such as to help indicate

the start of a new mail message in a list of mail messages.

CHAPTER 7

7.7

DIRECT CONTROL OF DURATION AND PITCH
Displaying the correct emotion through voice alone is a difficult task, as any

radio actor will tell you. The best method is to experiment with phonemic
symbols until you achieve the quality you want. Emotional content is usually
connected to the sentence content, so varying both together is the best way to

convey feelings.

For example, you can have DECtalk say a simple phrase like “Good morning”’
in several different ways.

Good morning.
(Normal tone.)

Good morning!
(Emphatic.)

Good morning?

(Questioning.)
[g"'uhd] morning.
(Emphasize ‘‘good.”)

If these alternatives do not produce what you need, you can use direct prosodic control. You must represent the entire sentence phonemically, specifying a
duration for each phoneme that does not match the natural model. You should

also give some or all phonemes specific target pitch values. DECtalk will

compute smooth transitions between pitch values, where the specified pitch is

reached at the end of the phoneme.

7.7.1

Duration and Pitch [<>]

DECtalk uses angle brackets [<>] to enclose duration and pitch values of
phonemes.

The format is

<duration,pitch>
where duration is the length of the phoneme in milliseconds (ms) and pitch is
the fundamental frequency of the phoneme in hertz (Hz).

DECtalk PHONEMIC INPUT

Any phoneme may be followed by angle brackets to alter the default duration
and pitch. If either value is omitted, or specified as 0, the default value is used.
The values for duration and pitch are separated by commas.

[ow]

(Normal phonemic specification.)

[0W<:1,00:»]

(1,000 ms duration.)
[ow<,90>]

(Default duration, 90 Hz pitch at end.)

[ow<1000,90>]

(1,000 ms duration, 90 Hz pitch at end.)

For example, to say “Oh?"” with a greater degree of skepticism than DECtalk
normally imparts, you could type

[<,90>0w<400,150>].

The [ow] phoneme begins at 90 Hz and ends (after 400 ms) at 150 Hz.
Note the use of the silence symbol [_] in the example just given. Pitch and

duration values must always be attached to a preceding phoneme.
The silence
symbol is used so that the value (90 Hz in this example) is applied to the
beginning tone of the next spoken phoneme [ow].
Many of the phonemes in Tables 7-1 and 7-2 (all except the stop consonants p,
t, k, b, d, and g) can be sustained in a monotone for an arbitrarily long duration
by using direct prosodic control. For example, to sustain “‘ah’ for a duration of
10 seconds (10000 ms) at a pitch of 120 Hz, type
[<,120>ah<10000,120>].
(Produces ‘‘ahhhhhhh - - -")

To produce a prolonged sigh, type

[<100,150>ah<2500,80>].

where the silence phoneme causes the pitch contour to start at 150 Hz at the
beginning of the ‘‘ah” and end at 80 Hz at the end of the ““ah.”
99

CHAPTER 7

7.7.2

Singing

Singing uses different voice control techniques than conversation. Even
untrained singers add liveliness to the sung notes by varying the pitch slightly,
a quality called vibrato. Singing in DECtalk sounds mechanical without vibrato.

Each word or syllable is defined phonemically. The first number following a
phoneme is the duration in milliseconds, and the second number is the pitch in

hertz. Vowels and consonants not assigned a pitch remain at the same pitch as
preceding segments. You can intersperse silence phonemes if you wish.

DECtalk stays exactly on pitch when the pitch is specified in hertz. You can add
vibrato (to give a more realistic singing quality) by specifying notes with pitch
values from 1 to 37. Note 1 is C2 and 37 is C5 on an equal tempered scale (A4
= 440 Hz) as shown in Table 7-4. C2 is the second C below middle C on a
piano, C4 is middle C, and so on.

When notes are specified, DECtalk reaches the desired pitch within about 100
ms after the start of the phoneme and adds vibrato after changing to this pitch.
When you give a specific nonsung pitch, DECtalk reaches the pitch target at

the very end of the phoneme with no vibrato. The following example makes
DECtalk “sing” the first four notes of Beethoven’'s Fifth Symphony.

[d<100,17>aa<400> d<100,17>aa<400>]
[d<100,17>aa<400> d<120,13>aa<700>].

100

DECtalk PHONEMIC INPUT

o
:

sl
5

e
M\N ;
o

L
e
e

é&
.

L
i

NoSio

L
"

o
i:

L
»}:z» -

e
i

.
-

q‘l

v%fi“ '
o

e
Gl
io

.
o

&:2

i
e

o
e

. .o

&Ns«w 1
deoh
.

. }:(.
.

(ARG

g;g‘

%%M

%}%?fl&%.

--=
--

a
.

""\3(
1
. y?:g%;@ .

.
.

i
b

s . s

o
o e
i

.
i

|
.

o ‘:;fifi

..
S

o ‘vfi‘ . .
i

.
o
.

101

CHAPTER 7

7.8

SELECTED READINGS

If you want to learn more about phonemics, acoustic phonetics, and speech
synthesis, the following references are suggested.
The Voice and Articulation Drillbook by G. Fairbanks, second edition, (Harper
and Row, 1960) is intended for speech therapists. This book includes
an
introduction to phonetics and acoustics. One of its diagnostic tests, the amplifi-

er passage (p. 114), was used to adjust DECtalk’s speech rate parameters.

The linguist responsible for DECtalk describes its operation in “The
Klattalk
Text-to-Speech System.” This article by D. H. Klatt comes from the 1982
proceedings of the International Conference on Acoustics, Speech, and

Processing (IEEE Catalog No. 82CH1746-7, pp. 1589-1592).

Signal

The Journal of the Acoustical Society of America, Volume 67, (Pp.
971-995)
has an article by D. H. Klatt. The article, ‘“‘Software for a Cascade/Paralleled
Formant Synthesizer,” describes the speech synthesizer that is the model
for
the one used in DECtalk.

For information on several speech synthesis technologies, read ‘‘Review

of the
Science and Technology of Speech Synthesis” by D. H. Klatt. This article
comes from the proceedings of the 1981 Meeting of the National Academy

Press in Washington, D.C., Overviews of Emerging Research Technique
s in
Hearing, Bioacoustics, and Biomechanics (pp. 15-33). (See also Speech
Group
Working Papers, MIT Research Laboratory of Electronics, Volume l1.)
g

102

You can modify the DECtalk voices by using phonemic commands and the
phonemic alphabet. This chapter describes a series of escape sequences that
gives a host computer slightly greater control over DECtalk. For example,
escape sequences can turn the DECtalk voice on or off and load the user
dictionary.

There are three ways to control DECtalk speech.

1. Through English text (sent@nws in standard English format and spelling).
DECtalk speaks this text as written.

2. Through phonemic spalling (sentences or phrases written in phonemic

symbols). Phonemic spelling is closer to the actual pronunciation of the

text.

3. Through phonemic commands. Phonemic commands control features of
speech that are not obvious from the visible text, such as rate of speech,
sex of the speaker, and excitement level.

103

CHAPTER 8

8.2

SPEECH TIMEOUT

Usually, DECtalk does not begin speaking until the host computer sends
a
clause terminator (period, comma, exclamation point, or question mark);
how-

ever, there is a 5-second timeout limit. If the host computer does not send data

within 5 seconds, DECtalk speaks the pending text in its input buffer, as ifa

(011) character had been sent.

Programs with long interruptions (such as pauses to search a
data base)
should collect complete sentences before sending anything to DECtalk. Other-

wise, this timeout may cause unnatural breaks in sentences and

ing speech.

8.3

jerky-sound-

ENGLISH TEXT

DECtalk speaks sentences written in standard English, if the text follows

rules.

three

1. Sentences end with a period, exclamation point, or question mark.
2. All commas, periods, exclamation points, and question marks are followed by a space (or an equivalent character from Table 3-1).

3. A period must be followed by enough text to distinguish between abbrevi-

~ations and the end of a sentence.

The host computer can send

English text in paragraph format; that is,

sentences can be broken in the middle by carriage returns.

If a sentence is too long to store in DECtalk’s buffers, the sentence is
spoken in
sections. DECtalk breaks up the sentence and speaks it as if clause boundaries were present. The effect is similar to a person trying to speak
a long

sentence and running out of breath. Keep sentences down to a
reasonable

length or insert commas to avoid this effect.

Chapters 7, 10, and Appendix C provide more information on speech phrasing

and emphasis. Section 8.6 in this chapter also describes how to coordinate
speech and interaction commands to prevent loss of information.

104

VOICE COMMANDS AND THE USER DICTIONARY

8.4

SPEAK PHONEMIC TEXT (DT_PHOTEXT)

When MODE_SQUARE is on, you can enter the phonemic mode and embed
phonemic twt in normal text by using the square brackets. When sending data
from the hqst computer, you can enter the phwnemtc mode by using the
DT_ PHOTEXT escape sequence as well as the square brackets;
MODEWSQUARE does not have to be on. The DT_PHOTEXT escape
sequence is as follows.

ESC

027

080

048 059

048

122

text

****

ESC

027

092

ESC P 0 : 0 z is the same as a left bracket ([), and ESC \ is the same as a right
bracket (]). DECtalk uses phonetic speech for all text between the command
terminator z and sequence terminator ESC \.

Appendix E of this manual lists the phonemic alphabet used by DECtalk.
In addition to sending the correct pronunciation, phonemic text can send control phonemes. This example changes the speech rate to 250 words per
minute.

ESC P
8.5

ESC \

:ra250

STOP SPEAKING (DT_STOP)

This escape sequence immediately stops speech, even uf DECtalk is in the
middle of a sentence. Any pending but unspoken text is lost, including index
markers that may have been sent by the host computer.

The DT_STOP escape sequence is as follows.

ESC P
027 080

0
048

;
059

1
049

0
048

z
122

ESC \
027 092

The stop sequence acts as a clause boundary. Speech stops immediately and
all internal buffers are reinitialized.

105

CHAPTER 8

8.6

DATA SYNCHRONIZATION (DT_SYNC)

The application program can send data to DECtalk faster than DECtalk can
speak it. If the user must carry on a dialogue with the application program

(through the telephone keypad), the application program should know whether

DECtalk has finished speaking the text sent to it. DT_SYNC provides this
coordination between the application program and DECtalk speech.
When the host computer sends DT_SYNC, DECtalk finishes speaking any
pending text before processing the next command from the host computer.

Therefore, the user hears a message before any other action starts, such

as
hanging up the telephone or starting the telephone timeout clock. Note
that
DT_SYNC acts as a clause boundary, the same as a comma, period, exclama-

PN,

tion point, or question mark.

The DT_SYNC escape sequence is as follows.

ESC

;

ESC

027

080

048

059

049

122

027

092

DT_SYNC does not reply to the host computer when processing is complete.

However, you can arrange to get a reply by following the DT_SYNC command

with a DT_INDEX_QUERY command.

8.7

ENABLE OR DISABLE SPEAKING (DT_SPEAK)

The DT_STOP sequence stops speech in progress. The DT_SPEAK sequence

turns speech processing off or on, so received text is either spoken or
discarded. DT_SPEAK is useful if the host computer can recognize such things
as
electronic mail letterheads. The host computer can act as a filter, removing

extraneous speech.

The DT_SPEAK escape sequence is as follows.

106

ESC

;

027

080

ESC

048

059

049

050

059

***

122

027

092

VOICE COMMANDS AND THE USER DICTIONARY

If P3 is 0, DECtalk stops speaking text; that is, it stops passing characters
received from the host computer to the text~tm~sp&ech processing section. If
P3 is not 0, ,DECtak resumes speaking.

DECtalk also resumes speaking if the host computer sends DT_SYNC,
DT_STOP, RIS, DECSTR, or any telephone command (for example
DT_PHONE:PH_ANSWER).

8.8

INDEXING

Text sent to DECtalk can contain index marks. DECtalk remembers these
marks when they are spoken. The host application can listen to the spoken text

(by reading the value of the last index) to determine how much transmitted text
was actually spoken. Index markers are truly marks; they do not modify
heuristics.
T

The following paragraphs describe how to mark text and return their values to
the application program.

8.8.1 Index Text (BT......INDEX)
The DT.JNIEX seqmnce inserts an index marker (flag) in the text stream sent
to DECtalk.

ESC
P

;

027

048

059

050 048

080

ESC

059

122

027

092

The P3 ,payragmet@r may range from 0 to 32767, sent
as the ASCII characters for
the number.@g Numbers outside the range are brought into range by unconditionally masking off any high order bits.

For example, the host computer sends the following data stream to DECtalk

and marks the first variable with the index 15.
Hello

ESC

;

ESC

there.

After speaking the text before DT_INDEX, DECtalk remembers the value 15.
The host computer may use DT_INDEX_QUERY (Section 8.8.3) to get this
stored value.

107

CHAPTER 8

8.8.2

Index Reply (DT_INDEX_REPLY)

DT_INDEX simply marks a position in the text. DT_INDEX_REPLY marks a
position, but also has DECtalk inform the host computer when the index is

spoken.
The DT_INDEX_REPLY escape sequence is as follows.

ESC

;

ESC

027

080

048

059

050

049

059

***

122

027

092

The P3 parameter is in the range 0 to 32767, using ASCII characters for the

selected number.
When DECtalk speaks the DT_INDEX_REPLY sequence, it sends a reply

(containing the P3 parameter of the index) to the host computer. The escape
sequence reply format is as follows.

ESC

;

ESC

027

080

048

059

051

049

059

***

122

027

092

P3 has the original value specified in DT_INDEX_REPLY.

8.8.3

Index Query (DT_.INDEX_QUERY)

DT_INDEX_QUERY requests DECtalk to reply to the host computer with the
last index marker spoken (that is, the last portion of spoken text that had an
index marker). The DT_INDEX_QUERY escape sequence is as follows.
A

ESC

;

ESC

027

080

048

059

050

122

027

092

DECtalk immediately returns a DECtalk reply escape sequence to the host
computer in the following format.

108

ESC

;

ESC

027

080

048

059

051

050

059

***

122

027

\
092

VOICE COMMANDS AND THE USER DICTIONARY

P3 contains the last index spoken. The P3 value is 0 under any of the following
conditions.

e The last index passed was 0.
e No index has been passed since power-up, RIS, or DECSTR.

Figure 8-1 shows how DT_SYNC and DT_INDEX_QUERY can coordinate
host-DECtalk communications.

HELLO. | AM DECtalk

| AM A SPECIAL COMPUTER
DEVICE THAT IMITATES
‘
HUMAN SPEECH.

HOST COMPUTER SENDS TEXT AND
~ DECtalk BEGINS SPEAKING.
...1 AM DECtalk
| AM A SPECIAL ...

ESCPO;11zESC
s

HOST COMPUTER SENDS DT_SYNC
ESCAPE SEQUENCE. DECtalk WILL

'NOT PROCESS ANY FURTHER
COMMANDS UNTIL IT FINISHES
SPEAKING THE CURRENT TEXT.

..COMPUTER DEVICE
THAT IMITATES

ESCPO0;22zESC \

HOST COMPUTER SENDS DT_INDEX_
QUERY ESCAPE SEQUENCE. DECtalk
' STORES THE SEQUENCE BUT DOES
NOT PROCESS IT.

ESCPO0;32;02ESC \
DECtalk FINISHES SPEAKING.

Figure 8-1

DT_INDEX_QUERY IS PROCESSED AND
' DT_QUERY_REPLY IS SENT TO HOST
COMPUTER. HOST PROGRAM NOW KNOWS
DECtalk HAS FINISHED SPEAKING.

MA-0356-85

Using DT_SYNC and DT_INDEX_QUERY to Coordinate DECtalk

Communications

109

CHAPTER 8

8.9

LOAD DICTIONARY (DT_DICT)

The user dictionary is used for processing abbreviations, and for providing
phonemic equivalents of unusual words. The DT_DICT escape sequence is as

follows.

ESC

;

name SP

027

080

phonemes ESC

048

059

052

048

122

****** (32

**wwxxxwxxsxs ()07

()92

Whenever the word represented by ‘‘name’” appears in the input text, the
phonemic pronunciation given by ‘“‘phonemes’ is used.
Any uppercase characters in the name only match uppercase characters in the

input text. Lowercase characters in the name match both uppercase and
lowercase characters in the input text. DECtalk always searches dictionary
entries in the order entered.
If a name ends with a period (.), a period must follow the word in running input
text. This period is included as part of the word, and is not recognized as a
sentence terminator.

Here are some examples of dictionary entries.
ESC

;

ms m’ihz ESC \

ESC

;

ms. m'ihz ESC \
DEC d’ehk ESC \

ESC

;

ESC

;

ESCP

dec dihs’ehmbrr ESC \

;

ESC

Goethe g’owth
iy ESC \

;

GOSLOW :ra 120 ESC \

;

Note the use of capitalization in the previous examples to distinguish between

abbreviations with the same spelling.
If you do not enter a phoneme substitution, DECtalk removes the word from
the user dictionary. You cannot remove words from the built-in dictionary.
After loading the word and its definition, DECtalk replies with a dictionary
status report as follows.

110

ESC

;

ESC

027

080

048

059

053

048

059

***

122

027

092

SIS,

VOICE COMMANDS AND THE USER DICTIONARY

P3 may have; one of the following values.
0

- The word was entered correctly.

048

No room is in the dictionary.

049

The entry is too long (256 characters maximum).

050

8.10 INPUT LOGGING (DT_LOG)
The DT_LOG command controls DECtalk logging of input text. This command
allows DECtalk to send the phonemes corresponding to the input text back to
the host computer.

ESC P
027 080

0
048

;
059

8
056

1
049

;
059

P3
***

059

P4
***

z
122

ESC \
027 092

The P3 parameter is interpreted as a set of flag bits, with only one bit
(LOG__HOSTPH, the bit with value 256) used. All other bits are ignored. The log

mask is set to 0 at power-up.

If the LOG__HOSTPH bit is set (P3 = 256) DECtalk will send the phonemic
representation of each clause (in ASCI| ‘characters) back to the host computer
as it is generated. The standard DECtalk phonemic alphabet is used. The last
logged phoneme will always be the clause delimiter. The clause delimiter will
always be followed by a carriage return.

The host computar must be prepared to receive the characters when they are
sent, or they will be lost. There is no flow control at all.

The P4 parameter is a level 2 device function. It controls the interpretation of
the P3 parameter as follows.

If P4 is 0, then P3 is a new log mask.

If P4 is 1, then P3 is a set of bits to be set in the log mask.

If P4 is 2, then P3 is a set of bits to be cleared in the log mask.

MODIFYING THE VOICES

This chapter shows how to change and create voices spoken
by DECtalk.
Changing and creating voices requires a certain knowledge of acoustic
phonetics and the human voice. This chapter necessarily goes into detail
on speakerdefinition parameters. Understanding the information is necessar
y for effective

voice modification and will make the task easier.

9.1 VOICE CHARACTERISTICS
The DECtalk DTCO03 system has a set of simple commands

that you can use to
change the speaking rate, or to change the voice to one of nine
different voices
- male, female, child, or a user-definable voice - as shown in
Figure 9-1. You

can use other, more complex commands to modify the character
istics of each
voice, or to create a new voice or special effects. The complex
commands
require skill and experience to use effectively, but the simple
commands are
easy to use in normal DECtalk applications.

When using the telephone, we can usually tell whether the

is that of a man, woman, or child. Slight differences in voice

voice of a stranger

quality are characteristic of these different speakers. For example, women’s
and children’s
voices are usually higher pitched than men’s voices. The
size of the head and
vocal tract account for some of the differences. We also
notice that some
people speak more quickly or more clearly than others.

Chapter 7 described ways you can modify DECtalk pronunciation. This chapter

shows how you can change the DECtalk voice itself by selecting
rate, sex, and other voice characteristics.

112

the speaking

MODIFYING THE VOICES

DOCTOR DENNIS

PERFECT PAUL

UPPITY URSULA

BEAUTIFUL BETTY

Figure 9-1

VARIABLE VAL

B4
HAM-OLGA-

The DECtalk Voices

113

CHAPTER 9
SR

DECtalk has five commands that you can use to modify voice characteristics

(Table 9-1). Because the commands are entered within phonemic brackets
and ], you must have MODE_SQUARE set. This option is set on at power-up.

MODE_SQUARE can also be set with the DT_MODE control sequence (Sec-

tion 4.6.2.)

NOTE: MODE_SQUARE is provided as a default. DECtalk interprets text
between square brackets as phonemes. For DECtalk to interpret the [ and ]
and the characters between them as normal text, and not the phoneme delimiters, MODE_SQUARE must be cleared.

WARNING: If you forget the final ], DECtalk will try to interpret ASCII text as
phonemes, skipping over illegal letter combinations. If DECtalk seems to
be

speaking text this way, type ] to recover.

The commands that modify the voice characteristics are

ARG

1. Speaking rate [:ra _]

2. Comma pause duration [:cp _]
3. Period pause duration [:pp _]
4.

New voice [:n_]

5. Design voice [:dv _]
o

where the “_"" represents a variable letter, value, or parameter.
s,

Each of the first three commands has a single, simple function. The fourth (new

voice) command selects the standard DECtalk voices. The fifth

design voice

command allows you to create a completely new voice.

114

MODIFYING THE VOICES

9.2 SPEAKING RATE [:ra ]

The default speaking rate is 180 words per minute (wpm). Speakmg rate values
have been c,ahbrated with a 300-word standard paragraph.” Speaking rates can
be adjusted to be very slow, very fast, or anywhere in between by using the
following commands.

This rate is the slowest, 120 wpm. It is ideal for situations

[:ra 120]

where material such as a phone number is to be copied
down by the listener.

WARNING: It is frustrating to listen to extended speech at slow rates unless
the listener is actually copying down every word.

This rate is moderate, 160 wpm. This rate sounds a little
slow, but may be preferred for telephone communication
because the reduced bandwidth of the telephone channel
makes human speech slightly less easy to understand than

[:ra 160]

in direct-contact communication.

[ra180]

[:ra 240]

This rate is normal (moderately fast), 180 wpm. It is the

default rate for DECtalk, and is ideal for listening to continuous text under optimal conditions.

This rate is fast, 240 wpm. Practiced listeners can skim mate-

rial at this rate and prefer it when scanning text for important
sections. Inexperienced listeners may not understand every
word at this rate.

[:ra 350]

This rate is the fastest, 350 wpm. It is too fast for most
people to follow, but it may have applications in special
circumstances.

Any speaking rates between 120 and 350 are permitted in the [:ra _] command.
Rates specified outside this range are limited to the nearest legal value.

Fairbanks, G. Voice and Articulation Drill Book. Second Edition. Harper and Row,
p. 114.
1960,

115

AR,

CHAPTER 9

Changes in speaking rate influence the duration and especially the number of
pauses in text, as well as the duration of individual phonemes. At rates below

140 wpm, DECtalk inserts pauses at all phrase boundaries. At rates faster than
240 wpm, DECtalk deletes the comma pause. At slow rates, pauses and phonemes near the ends of phrases are lengthened considerably. At fast rates,

these pauses and phonemes are shortened. (Near the beginning of phrases,
phonemes are fairly short at both slow and fast speaking rates.)

9.3

PAUSE DURATIONS [:pp —] and [:cp _]

At the normal speaking rate of 180 words per minute, DECtalk pauses about
half a second after a period in the text and about a sixth of a second after a
comma. These pause durations are adjusted appropriately when you change

the speaking rate.
In some situations, you might like a pause after a period without changing the
speaking rate. For example, to get DECtalk to read a list of words at a normal
rate with 5-second pauses after each word (to allow the listener to write them

down), you could use one of the following commands and end each word with
a comma (continuation rise intonation) or a period (falling intonation).

[:pp 4500]

Add a period pause of 4500 ms to the standard half-second

pause that occurs after a period in the text. The total pause

between words will be 5 seconds.
[:cp 4830]

Add a comma pause of 4830 ms to the standard sixth of a

second pause that occurs after a comma in the text at normal
speaking rate. The total pause between words separated by

a comma will be 5 seconds.

[:pp 0 :cp 0]

Reset the period pause and comma pause to their normal
default values.

The permitted range for a period pause is from -380 to 30000 ms. A negative
value shortens the standard period pause. The permitted range for a comma

pause is from —40 to 30000 ms. Values specified outside this range will be

limited to the nearest legal value.

116

MODIFYING THE VOICES

9.4 SELECTING A STANDARD VOICE [:n_]
DECtalk has nine built-in voices and one voice that is user definable. You can
refer to each voice by the command [:n_] where ‘" is a letter representing
one of the DECtalk voices. Table 9-2 lists the [:n_] symbols.
i

You can change voices with the new voice command as in this example.
[:nb] Hello. I'm Betty.

You can also change voices in the middle of a sentence.
[:np] This is a demo [:nb] of a sudden change in voice.

&4‘@4 .

R
5

ZH;
%

‘

;

o
s

e
s

o
e
.

e
.

g
.

L
o

s
0

|
.

S
g@}w,{.

Faine
o .

o
.

oo
o
A i
i

/\\{/w

m%:

i .

o
i

5
D

o S

.
L
7

.}?’3
.
4
L

. .

. ,,f 2

va a‘ .

}%fl‘"fiv
o

o
o Jf%&g{"

;

SWW
-

e
s
o

)

“

117

CHAPTER 9

DECtalk needs a brief silence to change
to a new built-in voice. If a voice
change request occurs in the middle of a sentence, DECtalk automatically
pauses slightly. The pause is the equivalent of inserting a comma before the

midsentence command. For example, you could type the previous sentence as

follows.

[:np] This is a demo, [:nb] of a sudden change in voice.

It is good practice to always end a sentence (insert a period) before changing
voices. This allows the listener to prepare for a new speaker.

9.5

DESIGNING A VOICE WITH SPEAKER-DEFINITION
PARAMETERS [:dv _]

The DECtalk voices provide an adequate selection for most user’s applications. However, if you have a special application requiring a monotone or
unusual voice (for example, the voice of a vocally handicapped person), you

can modify the parameters defined in this section on a trial-and-error basis to

get the desired voice.

The nine built-in voices of DECtalk are distinguished from one another by a

large set of speaker-definition parameters. Table 9-3 lists these parameters
and their ranges. Table 9-4 lists the parameters for all DECtalk voices.

Speakers can differ in sex, age, head size and shape, larynx size and behavior,

pitch range, pitch and timing habits, dialect, and emotional state. DECtalk

cannot approximate all of these options. Therefore, the space of distinguish

ble voices is quite limited, even though DECtalk has many speaker-definition

parameters that can be modified.
A

The design voice [:dv _] command introduces the speaker-definition parameters that can be entered as a string or one at a time.

The following sections discuss speech production, acoustics, and perception.

Some of the information may be difficult to follow, but the examples should

make it possible for all users to effectively modify any parameter and listen to

the results.

118

s
i

MODIFYING THE VOICES

;

: SRy
.

: ,

o
>

o
i

“

e
L.

,
o

.
i

e
e

@’?&
e

.
e

. %”f e
.

Vi
s .

‘

A
L

:
.

o
0

o
o

‘

e
!

s
e

o ”.(
.

.
L

.
.

G
e

‘M‘.‘u i

o
.

i
b

L
G

;

Ll
;

‘Ml"%
e

g
. g

‘i

‘)

wrJi

, S,

a0
,fi@i L
i J‘g ¢

o pi

119

CHAPTER 9
A,

S i

Gt
e
i

v
i
e

)

e
o

i T

;

e
e

- %*%

—
3
.
..

.
4

L
2

WM%WZ’?& 0

)

.
%\»fi%fi
i
e R S

o ’;fi i

gt
o

@4 e

~
s

el
.
g% .
5

)

hraan

o
.
o
M
Bt
R
o

)

,;@,

s
m;

Afi%

o
o

L f?""‘% L

Le g

i A ',fiafi"&

‘

i %’

g&%

fi%

& ?.

5
:

Giee

-~ o

]

;

)

o
S

o
o

.
»g-ff

)
o

o- S
-

o‘

“

‘fi}’

%M’?@

SeR

e
e
o

SN
e
e
e
oL
L ;‘Vfi%‘, e
s
G

e
.

.
e

;

S
S
R
.
e
.
e %%‘W
R
.
.
%"Q'w

’

= -~
-

G
S
o
e
e
e
L

o
o
M

a;\)"

e
o

| .

5 “s
-

e e
Gl
o

hdE

-o

e
e
o

a
Jlnae
P

ge
%%

;

e
L

)

‘

S
L

.
o
-

L
e
A

%@\ .

‘

il
R

A
"

s
o

;

=
o

0
)

“’fi o

¢
s
e

—

%fi=

.
:
-

§5

pess‘s%

" muj,?;%w\

L
ik

.
fl

‘

e
A
s

et
-

= -~

Sase

o
v

.:3

e
e

‘%, i

- =

o i

Ean

o
A

e (i

.
e

‘

= ':‘ KR

= *’%fig«zfié

P
=

g?%‘;?

e= %

N
5

i
.

= er;;_{

e
L e

e
G

L -

o o‘:n'
Z&‘
Lol

e
o

:«;3;.%@

o
=

A
i

e
SRR
.

»;i:

= 2o

‘

;

-~oo
=

5::(‘

o =

Mép}

—

e
i

*‘%;«., .

RE
SR
:S i

S gfitk%«% ’/),,@‘é‘:“';fa« s

T
G
e

e
S

‘*%

o
o

S
;

i
b

‘

Sh G

—

.
i

)

M@%fi

..
oo
S

i
.
e

e
.
:
;
e
¢
SR
g

.
W
T

h st"é\}i B
S
L
SR
.
a5
e
i
sk
i
2
e

i
A
ShEpe
s
N

120
EL

MODIFYING THE VOICES

’”«@afi o
i

9 i
o

o
‘
o
=

G
.

A‘E

.
Gy

%’tl”ii

s
e

L
o

f’fi%%
.

aen

%‘C

“‘W ;

v @*fi

“mfi%"v«
S

121

CHAPTER 9

9.5.1

Changing Sex and Head Size
Six speaker-definition parameters control the size and shape of the head.
These parameters are listed below and are described in Sections 9.5.1.1
through 9.5.1.3.

Sex,

hs
f4

Head size, in percent
|
Fourth formant resonance frequency, in Hz

Fifth formant resonance frequency, in Hz

morf,1or0

Fourth formant bandwidth, in Hz

Fifth formant bandwidth, in Hz

9.5.1.1

Sex, sx - Male and female voices have many differences, including

head size, pharynx length, larynx mass, and speaking habits such as degree of

breathiness, liveliness of pitch, choice of articulatory target values, and speed
of articulation. Some of these differences are under the control of a single

parameter, sx, the sex of the speaker. Speakers Paul, Harry, Frank, and Dennis are male (sx = 1), while speakers Betty, Rita, Ursula, Wendy, and Kit are
female (sx = 0). Actually, Kit the Kid can be male or female because children

younger than 10 years old have similar voices for both sexes.

Changing the sx parameter causes DECtalk to access a different (male or
female) table of target values for formant frequencies, bandwidths, and source

amplitudes. The male and female tables are patterned after two individuals who
were judged to have pleasant, intelligible voices. DECtalk’s built-in voices are
only scaled transformations of Paul and Betty, the two basic voices.

You can change the sex of any of DECtalk’s voices by making the voice current

and then modifying the sx parameter. For example, the following command
gives Paul some of the speaking characteristics of a woman. (The sx parameter does not change the average pitch or breathiness, so a peculiar combination of simultaneous male and female traits results from this sx change.)
[:np :dv sx 0] Am | a man or woman?

122

MODIFYING THE VOICES

The sx parameter can also be specified as m orf with the commands [:dv sx m]
or [:dv sx f].

WARNING:

If you change the sex of the voice, some phonemes may cause

DECtalk’s filters to overload, producing a squawk. (The squawk is unpleasant,
but it will not damage DECtalk.) Sections 9.5.1.3 and 9.5.4.3 describe parameters f4, 5, and g1 and explain how to correct this problem.

9.5.1.2

Head Size, hs - Head size is specified as the average size for an adult

man (if sx = 1) or an adult woman (if sx = 0). A head size of 100 percent is
normal or average for a given sex, but people can differ quite a bit in this
characteristic. Head size has a strong influence on a person’s voice. Large
musical instruments produce low notes, and humans with large heads tend to
have low, resonant voices. For example, to make Paul sound like a larger man
with a 15 percent longer vocal tract (and formant frequencies that are scaled
down by a factor of about 0.85 percent), type the following command.
[:np :dv hs 115] Do | sound more like Huge Harry this way?

Head size is one of the best variables to use if you want to make dramatic
voice changes. For example, Paul has a head size of 100, while Harry’s deep
voice is caused in part by a head size change to 115, or 15 percent greater than
normal. Decreasing head size produces a higher voice, such as in a child or
adolescent. Extreme changes in head size, as in the following examples, are
somewhat difficult to understand.

[:nh :dv hs 135] Do | have a swelled head?
[:nk] | am about 10 years old.

[:nk :dv hs 65] Do | sound like a six year old?

WARNING: Extreme changes in head size can cause overloads, as well as
difficulties in understanding the speech. To correct an overload, see Sections
9.5.1.3 and 9.5.4.3 on parameters f4, 15, and g1.

123

CHAPTER 9

9.5.1.3

Higher Formants, f4, f5, b4, and b5 - A male voice typically has five

prominent resonant peaks in the spectrum (over the range from 0 to 5 kHz), a

female voice typically has only four (due to a smaller head size), and a child has
three. If fourth and fifth formant resonances exist for a particular voice, they
are fixed in frequency and bandwidth characteristics. These characteristics are
specified by the parameters f4, f5, b4, and b5, in Hz. Values for each
predefined voice are in Table 9-4.

If a higher formant does not exist, the frequency and bandwidth of the speaker
definition are set to special values that cause the resonance to disappear. To
make a resonance disappear, the frequency is set to 2500 Hz, and the
bandwidth to 2048 Hz. This is what has been done to the fourth and fifth

formants for Kit the Kid (Table 9-4).
The permitted values for f4 and f5 have fairly complicated restrictions. Violating
these restrictions can cause overloads and squawks. The restrictions are listed

below for cases where a higher formant exists.
1.

F5 must be at least 300 Hz higher than f4.

2. If sx is male, f4 must be at least 3250 Hz.

3. If sx is female, f4 must be at least 3700 Hz.
4.

If hs is not 100, the above values should be multiplied by (hs / 100).

These higher formants produce peaks in the spectrum that become more
prominent if b4 and b5 are smaller, and if f4 and 5 are closer together. The
limits placed on b4 and b5 (Table 9-3) should ensure that no problems occur.

However, smaller values for bandwidths may produce an overload in the syn-

thesizer. You can correct these overloads by increasing the bandwidths or by
changing the gain control g1 (Section 9.5.4.3).

]

124

MODIFYING THE VOICES

Changing Voice Quality

9.5.2

Six speaker-definition parameters control aspects of the output of the larynx,
which, in turn, control voice quality. These parameters are listed below and are
described in Sections 9.5.2.1 through 9.5.2.6.

Breathiness, in decibels (dB)

Lax breathiness, in percent

Smoothness, in percent

Richness, in percent

nf
la

Number of fixed samples of open glottis
Laryngealization, in percent

9.5.2.1 Breathiness, br - Some voices can be characterized as breathy. The
vocal folds vibrate to generate voicing and breath noise simultaneously.
Breathiness is a characteristic of many female voices, but it is also common
under certain circumstances for male voices.

The range of the br parameter is from 0 dB (no breathiness) to 70 dB (strong

breathiness). By experimenting, you can learn what intermediate values sound
like. Noting breathiness values in Table 9-4 for each of the predefined speakers
may also help. For example, to turn Paul into a breathy, whispering speaker,

type the following command.

[:np :dv br 55 gv 56] Do | sound more like Doctor Dennis now?

This voice is not as loud as the others due to the simultaneous decrease in the
gain of voicing, gv, but it is intelligible and human sounding.

9.5.2.2 Lax Breathiness, Ix - The br parameter (Section 9.5.2.1) creates
simultaneous breathiness whenever voicing is turned on. Another type of
breathiness occurs only at the ends of sentences and when going from voiced
to voiceless sounds. This type of “lax” breathiness is controlled by the Ix
parameter in percent.

A nonbreathy, tense voice would have Ix set to 0, while a maximally breathy,
lax voice would have Ix set to 100. The difference between these two voices is
not great, but you can hear it if you listen closely.

125

CHAPTER 9

9.5.2.3

Smoothness, sm - Smoothness refers to vocal fold vibrations. The

vocal folds meet at the midline, as they do in normal voicing, but they do not

slam together forcefully to create a very sudden cessation of airflow.

DECtalk uses a variable-cutoff, gradual low-pass filter to model changes to

smoothness. The range of sm is from 0 percent (least smooth and most

brilliant) to 100 percent (most smooth and least brilliant). The voicing source
spectrum is tilted so that energy at higher frequencies is attenuated by as

much as 30 dB when smoothness is set to a maximum, but is not attenuated at
all when smoothness is set to 0.
Professional singing voices that are trained to sing above an orchestra are
usually brilliant, while anyone who talks softly becomes breathy and smooth.

To synthesize a breathy voice, an sm value of about 50 or more is good.
Changes to sm do not have a great effect on perceived voice quality.

9.5.2.4 Richness, ri - Richness is similar to smoothness and brilliance,
except that the spectral change occurs at lower frequencies, and is due to a

different physiological mechanism. Brilliant, rich voices carry well and are more

intelligible in noisy environments, while smooth soft voices sound more friend-

ly. For example, typing the following command produces a soft, smooth ver-

sion of Paul’s voice.

[:np :dv ri 0 sm 70] Do | sound more mellow?
The following command produces a maximally rich and brilliant (forceful) voice.

[:np :dv ri 90 sm 0] Do | sound more forceful?

Smoothness and richness are usually negatively correlated when a speaker

dynamically changes laryngeal output. The sm and ri parameters do not influence the speaker’s identity very much.

126

MODIFYING THE VOICES

9.5.2.5 Nopen Fixed, nf - The number of samples in the open part of the
glottal cycle is determined not only by ri, but also by a second parameter nf.
NF is the number of fixed samples in the open portion of the glottal cycle

Most speakers adjust the open phase to be a certain fraction of the period, and
this fraction is determined by ri. Other speakers keep the open phase fixed in
duration when the overall period varies. To simulate this behavior, set ri to 100
and adjust nf to the desired duration of the open phase. The shortest possible
open phase is 10 (1 ms), and the longest is three quarters of the period
70 for a male voice).
duration (about

9.5.2.6 Laryngealization, la - Many speakers turn voicing on and off irregularly at the beginnings and ends of sentences, which gives a querulous tone to
the voice. This departure from perfect periodicity is called laryngealization or
creaky voice quality.

The la parameter controls the amount of laryngealization in the voice. A value
and a value of 100 (the maximum)
of 0 results in no laryngealized irregularity,
produces laryngealization at all times.

For example, to make Betty moderamly Iaryngaa ized, type the following
oommand

[:nb :dv la 20]

The la parameter creates a noticeable difference in the voice, although it is not
altogether a pleasant change.

127

CHAPTER 9

9.5.3

Changing the Pitch and Intonation of the Voice
Seven speaker-definition parameters control aspects of the fundamental fre-

quency (f0) contour of the voice. These parameters are listed below and are

described in Sections 9.5.3.1 through 9.5.3.6.

Baseline fall, in Hz

Hat rise, in Hz

Stress rise, in Hz

‘as

Assertiveness, in percent

Quickness, in percent

Average pitch, in Hz

Pitch range, in percent

9.5.3.1

Baseline Fall, bf - The bf parameter (baseline fall in Hz) determines

one aspect of the dynamic fundamental frequency contour for a sentence. If bf

is 0, the reference baseline fundamental frequency of a sentence begins at 115
Hz and ends at this frequency. All rule-governed dynamic swings in fO are
computed with respect to the reference baseline.

Some speakers begin a sentence at a higher f0, and gradually fall as the
sentence progresses. This “falling baseline” behavior can be simulated by

setting bf to the desired fall in Hz. For example, setting bf to 20 Hz will cause
the f0 pattern for a sentence to begin at 125 Hz (115 Hz plus half of bf), and fall
at a rate of 16 Hz per second until it reaches 105 Hz (115 Hz minus half of bf).
The baseline remains at this lower value until it is reset automatically before the

beginning of the next full sentence (right after a period, question mark, or
exclamation point). The rate of fall, 16 Hz per second, is fixed, no matter what
the extent of the fall.
Whenever you include a [+] phoneme in the text to indicate the beginning of a
paragraph, the baseline is automatically set to begin slightly higher for the first

sentence of the paragraph. The following sentences of a paragraph are all
identical in having a normal baseline fall.

While baseline fall differs among the speakers, it is not a very good cue for
differentiating between speakers. As long as the fall is not excessive, its presence or absence is not particularly noticeable.

128

MODIFYING THE VOICES

9.5.3.2 Hat Rise, hr — The hr (nominal hat rise in Hz) and sr (nominal stress
impulse rise in Hz) parameters determine aspects of the dynamic fundamental
frequency contour for a sentence. To modify these values selectively, you

should understand how the fO contour is computed as a function of lexical
stress pattern and syntactic structure of the sentence.
A sentence is first analyzed and broken into clauses with punctuation and

clause-introducing words to determine the locations of clause boundaries.
Within each clause, the fO contour rises on the first stressed syllable, stays at a
high level for the remainder of the clause up to the last stressed syllable, and
falls dramatically on the last stressed syllable. The rise-at-the-beginning and
fall-at-the-end pattern has been called the ‘‘hat pattern’ by linguists, using the

analogy of jumping from the brim of a hat to the top of the hat, and back down
again.

The hat rise parameter, hr, indicates the nominal height in hertz of a pitch rise

to a plateau on the first stress of a phrase. A corresponding pitch fall is placed
by rule on the last stress of the phrase. Some speakers use relatively large hat
rises and falls, while others use a local “impulse-like’’ rise and fall on each

stressed syllable. (See Section 9.5.3.3.) The default hr value for Paul is 22 Hz,
indicating that the fO contour rises a nominal 22 Hz when going from the brim to

the top of the hat. To simulate a speaker who does not use hat rises and falls,
enter the command [:dv hr 0].

Other aspects of the hat'patmm are important for natural intonation but are not
accessible by speaker-definition commands. For example, the hat fall becomes
a weaker fall followed by a slight continuation rise if the clause is to be
succeeded by more clauses in the same sentence. Also, if unstressed syllables
follow the last stressed syllable in a clause, part of the hat fall occurs on the

very last (unstressad) syllable of the clause. If the clause is long, DECtalk may
break it into two hat patterns by finding the boundary between the noun phrase
and verb phrase.

If DECtalk is in phoneme input mode and you use the pitch rise [/] and pitch fall

[\] symbols (Section 7.6.5), the hr parameter determines the actual rise and fall
in Hz.

129

CHAPTER 9

9.5.3.3

Stress Rise, sr - The sr parameter indicates the nominal height, in Hz,

of a local pitch rise and fall on each stressed syllable. This rise-fall is added to
any hatrise or fall that may also be present. (See Section 9.5.3.2.) For example,
Paul has pr set to 32 Hz, resulting in an f0 rise-fall gesture of 32 Hz over a span
of about 150 ms, which is located on the first and succeeding stressed sylla-

bles. However, DECtalk rules reduce the actual height of successive stress
rises and falls in each clause, and cause the last stress pulse to occur early so
that there is time for the hat fall during the vowel.
If the sr parameter is set too low, the speech sounds monotone within long
phrases. Great changes to hr and sr from their default values for each speaker
are not necessary or desirable, except in unusual circumstances.

9.5.3.4

Assertiveness, as - Assertive voices have a dramatic fall in pitch at

the end of utterances. Neutral or meek speakers often end a sentence with a
slight “questioning” rise in pitch to deflect any challenges to their assertions.
The as parameter, in percent, indicates the degree to which the voice tends to
end statements with a conclusive final fall. A value of 100 is very assertive,
while a value of 0 is maximally meek.

9.5.3.5 Quickness, qu - The qu parameter, in percent, controls the speed of
response to a request to change the pitch. All hat rises, hat falls, and stress

rises can be thought of as suddenly applied commands to change the pitch, but

the larynx is sluggish, and responds only gradually to each command. A smaller larynx typically responds more quickly, so while Harry has a quickness value
of 10, Kit has a value of 50.
In engineering terms, a value of 10 implies a time constant (time to get to 70
percent of a suddenly applied step target) of about 100 ms. A value of 90
percent corresponds to a time constant of about 50 ms. Lower quickness

values may mean that the fO never quite reaches the target value before a new
command comes along to change the target, but this is perfectly natural.

9.5.3.6

Average Pitch, ap, and Pitch Range, pr - The ap (average pitch in Hz)
and pr (pitch range in percent of normal range) parameters modify the comput-

ed values of fundamental frequency, f0, according to the formula:

fO’ = ap + (((fO - 120) * pr) / 100)

130

MODIFYING THE VOICES

If ap is set to 120 Hz and pr to 100 percent, there will be no change to the
“normal’’ f0 contour that is computed for a typical male voice. The effect of a
change in ap is simply to independently raise or lower the entire pitch contour
by a constant number of Hz, while the effect of pr is to expand or contract the
swings in pitch about 120 Hz.

Normally, a smaller larynx simultaneously produces f0 values that are higher in
average pitch and higher in pitch range by about the same factor (the whole f0
contour is multiplied by a constant factor). Observing the values assigned to ap
and pr for each of the voices (Table 9-4), you can see that the voices rank in
average pitch from low (Harry) to high (Kit). Rankings for pr are similar, except
e range compared with his average
pitch
that Frank has a flat, nonexpressiv
pitch.

The best way to determine a good pitch range fora new voice is by trial and
error. You can create a monotone or robot-like voice by setting the pitch range
to 0. For example, to make Harry speak in a monotone at exactly 90 Hz, type
the following command.

[:nh :dv ap 90 pr 0] | am a robot.

Reducing the pitch range reduces the dynamics of the voice, producing emotions such as sadness. Increasing the pitch range while leaving the average
pitch the same or setting it slightly higher suggests excitement.

in pitch-synchronous updating of other dynamically
Due to constraints involved
frequency contour that is computed by
fundamental
the
changing parameters,
the above formula is then checked for values that are out of bounds with
respect to the following limits.

f0 maximum = 500 Hz
fO minimum = 50 Hz

Any value outside this range is limited to fall within the range.

‘To keep you from exceeding masmabla limits on the parameters controlling

pitch, the constraints in Table 9-3 have been placed on values selected. If a [:dv
] commargd requests values outside these limits, the request is limited to the

nearest listed value before execution.

131

CHAPTER 9

9.5.4

Changing Relative Gains and Avoiding Overloads

Eight speaker-definition parameters control the output levels of various internal
resonators. These parameters are listed below and are described in Sections

9.5.4.1 through 9.5.4.3.
lo

Loudness of the voice, in dB

Gain of voicing source, in dB

Gain of aspiration source, in dB

Gain of frication source, in dB

Gain of cascade formant resonator 1, in dB

Gain of cascade formant resonator 2, in dB

Gain of cascade formant resonator 3, in dB

Gain of cascade formant resonator 4, in dB

9.5.4.1

Loudness, lo - Each predefined voice has been adjusted to have
about the same perceived loudness, a value that is about optimum for tele-

phone conversation. The value chosen is near maximum (if loudness were
increased much, some phonemes would probably cause an overload squawk).

A near maximum value was selected to maximize the signal-to-noise level of

DECtalk.

If you want to decrease the loudness of a voice, or make a temporary increase

for a phrase that is known not to overload, determine the lo value in dB for the

voice in question by referring to Table 9-4. Then adjust the voice by using the

following command.

[:np :dv lo 76]. | am speaking at about half my normal level.
Because the lo entry in Table 9-4 for Paul is 86, this command reduces loud-

ness by 10 dB. Perceived loudness approximately doubles (or halves) for each

10 dB increment (decrement) in lo.

Software control over loudness is useful in a loudspeaker application where

the background noise level in the room might change. For example, a vocally

handicapped wheelchair-bound person does not want to appear to be shouting

in a quiet interpersonal conversation, but may wish to be able to converse in a
noisy room as well. Using a software abbreviation facility, such a person could

type lo to select

a command making the voice maximally loud, and sof to
invoke a command setting lo to a reduced value.

132

MODIFYING THE VOICES

9.5.4.2 Sound Source Gains, gv, gh, and gf - Three types of sound sources
are activated during speech production: voicing, aspiration, and frication. The
relative output levels of these three sounds, in dB, are determined by the gv,
gh, and gf parameters respectively. The default settings (Table 9-4) for these
parameters have been factory preset to maximize the intelligibility of each
voice. However, changing the settings can be useful in debugging the system

or in demonstrating aspects of the acoustic theory of speech production. You
could change the level of one sound source globally, for example, turning off
frication can allow you to hear just the output of the larynx. These parameters
might have to be reduced to overcome certain kinds of overloads, but try the
procedure in the next section first.

9.5.4.3 Cascade Vocal Tract Gains, g1, g2, g3, and g4- Changes in head
size or other parameters can sometimes produce overloads in the synthesizer
circuits. If this occurs, first check to see that f4 and f5 are set to reasonable
values. If the squawk remains, you can adjust several gain controls, g1 through
g4 in dB, in the cascade of formant resonators of the synthesizer to attenuate

the signal at critical points. These gains can then be amplified back to desired
output levels later in the synthesis.

Use the following procedure to correct an overload (typically indicated by a
squawk during part of a word).

1. Synthesize the word or phrase several times to make sure the squawk
occurs consistently. Use the same test word each time one of the following changes to a gain is made.

2. See Table 9-4 to determine the default values for g1 through g4 for the
speaker that overloads.

3. Reduce g1 by an increment of 3 at a time until the squawk goes away.
When the squawk goes away, note the reduction that was needed. If
more than a 10 dB decrement is required, some other parameter has
probably been changed too much. If the squawk does not go away at all,
then you may need to reduce gv instead of g1.

4. Add this increment to lo to return the output to its original level. For
example, if g1 was reduced by 6 dB, add 6 dB to lo (or g4 if lo is already at
a ma;{ximum). If incrementing lo causes the squawk to return, then

decrease lo slowly until the squawk goes away.

CHAPTER 9

This procedure works in most cases, but using g2 rather than g1 can work
better. If you can return g1 to its factory preset value (Table 9-4) and reduce g2
instead to make the squawk go away, then the signal-to-quantization-noise
level in g1 remains maximized. If you can fix the squawk by using g3 or g4

rather than g2, more of the cascaded resonator system can be made immune
to quantization noise accumulation.

9.5.5

The [save] Parameter and [:nv] Voice

You can save a modified speaker definition in a buffer while synthesizing
speech with one of the other voices. The Variable Val voice [:nv] is either male

or female, depending on what values are stored in the buffer. If you call Val
before storing any values in the buffer, DECtalk uses the Perfect Paul voice

[:np]. The following commands store a modified Betty voice in Val and then
recall it.

[:nb :dv sex m save]
(Store the modified Betty voice in Val.)
[:np] | am Paul.

(Use another voice.)
[:nv] | am Val.
(Recall the Val [modified-Betty] voice.)

The buffer holds its contents until you power down DECtalk. You must reenter
new voice characteristics if you turn off DECtalk.

9.5.6 Summary on Speaker-Definition Parameters
Of the 27 parameters, only a few cause dramatic changes in the voice. The
greatest effects are obtained with changes to hs, ap, pr, and sx, while moder-

ate changes occur when modifying la and br. To some extent, DECtalk’s nine
factory-set speakers cover most of the possible voices, so don’t expect to be
able to find a voice that is highly novel and intelligible. However, you might

easily find ways to slightly improve one of the standard voices.
9.6

VOICE COMMAND SYNTAX

DECtalk uses the following voice command syntax rules.
1.

Begin every command with a colon (:).

2. Separate each command and its parameter(s) from the text by a valid
word boundary marker such as a space, tab, or carriage return.
134

iy
g

MODIFYING THE VOICES

. You can include several commands in the same square bracket set.
[:ra

150 :nb] Hello. How are you?

You can include several parameters in the same square bracket set if the

command allows more than one parameter. If you use several parameters, you must give them all before a second command in the same

square bracket set.
[:dv ap 160 pr 50 save :nv] Hi there.

(The parameter group modifies the [:dv _] command.)

[:dv ap 160 save :nv pr 50] Hi there.

(Wrong. The parameter group is out of place.)

. f youigive two conflicting parameters or commands, DECtalk will use the
last command in the sequence. For example, if you type
[:nb :np] Hello.
DECtalk will use Paul’'s voice.

. You can use phonemic symbols in the same square brackets with voice
commands.

Now I'm [:dv ap 90 pr 130 r”iyliy] thrilled!

. If the value in a [:dv _] command is too low, DECtalk will use the minimum
valid value. If the value is too high, the maximum valid value will be used.

Once you give a command, that command applies to all further text until
overridden by another command. For example, the command

will make DECtalk use Kit's voice on all entered text until you enter

another new voice command.

All [:dv _] commands are lost when you power down DECtalk.
10. Invalid commands are ignored.
135

10 HOW TO GET THE

MOST OUT OF DECtalk

This chapter gives basic rules on how to develop an application and examples

of how to use DECtalk in the most efficient way.
10.1

DEVELOPING YOUR APPLICATION

DECtalk lets people use computer-based applications from any keypad telephone. DECtalk speaks messages in an understandable voice. When the user
presses keypad keys, DECtalk sends characters to the host program. The

following guidelines are for adapting an application to unique customer needs.

General Guidelines
e Relate to the problem from the user’s point of view, not the programmer's. Use commands that are logically related to the way users see the

task.
e

136

Frequent users become experts quickly.

HOW TO GET THE MOST OUT OF DECtalk

”~

Writing Dialog

e Keep dialog simple, but meaningful,

* Organi:e each’message as follows.
1. Put the hardest elements to remember first.

2. Put the easiest elements to remember in the middle.
3. Put information for immediate recall at the end.

e Tell users only what they need to know to continue a task.

e Do not use humor or threats. Keep dialog strictly factual and informative.

For example, say ‘‘Please try again,” rather than “‘Let’s try again.”

Help Messages and Replies

e Make help messages optional. Let users decide when they want more
information.

e Repeat significant phrases in help messages.

e Let users know that DECtalk is acting on their specific commands. For
example, say ‘‘Sending reply to Ms. Jones,’’ rather than ‘Sending reply.”

Entering Keypad Commands

e Remember, a telephone keypad has only 12 keys.
-

e Keep the same function on the same key.

e Refer to keypad numbers, not letters. People do not remember which

X
o

letter is on which key. Use “Press 1 for next, 2 for previous, 3 to exit,”
rather than ““Press N for next, P for previous, E to exit.”

e Create a standard method for users to exit from a subtask to the main
dialog.

137

CHAPTER 10

10.1.1

Names, Part Numbers, and Alphanumeric Text

In many DECtalk applications, you use the 12 keypad keys to enter a person’s
name or an alphanumeric part number.

Because the application program

receives only a string of digits (and the # and * characters), the program must

use the digits as an index to the actual data item.
If you are designing a new system, you could specify numeric part numbers
only. However, in the real world, a company is not going to change its existing
warehouse methods to match DECtalk. The user will have to enter something

that your application can translate into the current system.

10.1.2

Direct Numeric Encoding

Using this method, the user simply presses the key labeled with the desired
letter. For example, to select “DIGITAL” the user would press 3444825. You
could assign the letters Q and Z to the 7 (PQRS) and 9 (WXYZ) keys,
respectively.
Numeric encoding is a simple method to describe and implement. Because
users can recall more than one item for a given digit string, your application
must provide a way to select alternatives. You could have users select alterna-

tives by number, or you could have them step through a list by using next and
previous commands.

Numeric encoding is probably the best method for lists of names and for many
part number applications. You can use this method for ID or password entry.

10.1.3

Two-Character Encoding

Some applications use specific letters in their codes (for example, three-character airport codes). You cannot use direct numeric encoding to select specific

letters on the keypad.

One possible solution is two-character encoding. This method matches the
three letters on each key to the three columns of keys on the keypad. The user
presses two keys to select a letter.

1. The key with the desired letter

2. The 1, 2, or 3 key (to select the specific letter)
For example, to select “DEC,” the user would press 313223. You could have
users enter numbers together with the 0 (OPER) key. And you could assign the

missing Q and Z (plus the space character) to the 1 key.
138

HOW TO GET THE MOST OUT OF DECtalk

Ending Commands and Data

10.1.4

P
=

You can use single-character commands and fixed-length data fields for many

applications. But for complex applications or variable-length data, you may find
it simpler to ask the user to end all commands and data by pressing a special
key (such as #). Pressing # lets the program know that the right number of
characters have been entered.
10.1.5

Application Development Tips

Here are some tips for encoding the application itself.

e Use timeouts for everything. Assume that the user may hang up the

phone at any time. Also assume that data entry will be quite slow. This is
important when planning data base entry and record-locking strategies.

e People can recall about 5 seconds of text without difficulty. You can use
entries such as “‘1 for yes, 2 for no, 3 for maybe,” but do not ask an

.
-

e DECtalk tends to spell out text that may be ambiguous (for example, part
numbers). You can write a small subroutine that recognizes certain

untrained user to remember anything more complex.

strings and pronounces them in a form more suitable for your specific

application.

e If your application accepts data from the telephone keypad, make sure

the operating system can buffer type-ahead characters. Also, make sure
the operating system responds to DECtalk’s XOFFs.

o DECtalk speaks pending text if the host computer stops delivering text for
5 seconds. This feature may be a problem on an overloaded system. You
may need help from the system manager to get more resources or adjust
program priorities.

e When you have DECtalk speak information from a data base, remember

repeat function for complex subject matter. If you have DECtalk read mail
or other unstructured text, you should offer a back up one sentence

been heard.

;

that the listener hears the information only once. You should offer a

function by using the index commands (Section 8.8) to signal what has

139

CHAPTER 10

10.2 DEVELOPING AN ADVANCED APPLICATION
The development process described in this manual assumes that your applica-

tion has full control over the text being spoken. If you are developing an

application that must read arbitrary text (such as electronic mail messages),
your task is more difficult because almost anything can appear in the text.

In most applications, DECtalk is controlled by a computer. Even the smallest
personal computer has enough power to preprocess (filter) text to handle
application-unique cases. So, you can put application-specific text filters in the
controlling computer,

rather than add many additional special cases (and

switches to enable and disable them) to DECtalk.

For an electronic mail system, you can program an electronic mail preprocessor to make the following text conversions before sending the text to DECtalk.

Parse the header boilerplate to remove extraneous information.

2. If DECtalk is speaking paragraphs of text, add the [+] symbol to a blank
line separating each paragraph.

3. If words are separated by / or another special character, replace the /
with a space (‘‘Raleigh/Durham’ should become ‘“Raleigh Durham’ for
DECtalk to say it without spelling out the entire string).
4. Write your own application-specific dictionary for words, such as proper
names, that DECtalk mispronounces. If DECtalk is connected to a data

base containing names, consider adding a pronunciation field to the name
record, entering phonemic text when appropriate.

140

HOW TO GET THE MOST OUT OF DECtalk

TM
-

5. Scan the text for strings of numbers in a format understandable to your
application but not to DECtalk. For example, if you can extract the time
format from an electronic mail message, you can add code to your appli-

cation to expand it to its “o’clockTM form.

6. In many applications, the listener will want to write down number strings

for strings of numbers and, when found, send them to DECtalk in a way

(such as prices or telephone numbers). Your application can scar the text
that includes pauses at critical locations. For example:
The number is, 1 (800) 555,12 3 4. [:ra 120]

That is, [_<300>] 1 (800), [_<500>] 5 5 5, [_<900>] 1 2 3 4. [:ra 180].

The spaces between the numbers ensure that ‘‘five five five'’ is spoken
rather than ‘‘five hundred and fifty five.”” The slower speaking rate and the
silence phonemes of specified durations were carefully selected to permit
enough time for the listener to write down the entire number. Silence
phonemes were positioned after an orthographic comma to minimally

o
-

disturb the intonation.

As another example, if your application speaks money (such as bank
balances or item costs), it might say

Your balance is $244.05

~
-

That is, 2 4 4, [_<400>] point 0 5, [-<400>] dollars.

7. When spelling an item out, your application may have to distinguish the
case of letters. Consider using different voices to distinguish between
uppercase and lowercase letters (for example, Harry and Paul).

141

CHAPTER 10

10.3 OPTIMIZING THE QUALITY OF SPOKEN TEXT
In some applications, it is important to get a few sentences to sound very good

because they are used often (for example, to open a dialog with a customer).

Usually DECtalk does a satisfactory job, but sometimes it mispronounces a
word or gets the phrasing wrong. In these cases, you may wish to spend the

time required to optimize the quality of a particular sentence. The following
steps are suggested.

1. Send the sentence to DECtalk and listen repeatedly, focusing on each
word in turn to detect any mispronunciations.
.

For each word that is mispronounced, there are several alternatives to get

the correct pronunciation.

For words that have two alternate pronunciations, and DECtalk has cho-

sen the wrong one, try placing a right parenthesis in front of the spelling,
which asks for the second pronunciation. (See Section 7.6.10 and Appen-

dix D, which contains a complete list of words for which two pronuncia-

tions are known to DECtalk.)

| )read yesterday that - - Replace the correct spelling of the word with a clever misspelling (Section

7.3.).

| red yesterday that - - -

If the word is a compound, use a hyphenated spelling to help DECtalk see
the two parts of the compound (Section 7.6.9.).
The slide-show host - - -

Replace the text version by a phonemic string. Use the commands and
phonemic symbols described in Section 7.5. Be sure to place the lexical
stress pattern correctly (Section 7.6). For example, you can replace the

word “Alphonse’ with [aelf’aans].

142

““““““““““““

HOW TO GET THE MOST OUT OF DECtalk

3. Sometimes, a word does not sound quite right even when the best alter-

nate phonemic representation is selected. Usually, such subtle pronunciation defects are not correctable. However, if they are due to incorrect
phoneme durations, then you can specify the duration
of each phoneme

with the commands described in Section 7.7.1. This process is difficult

and probably should not be tried until you have mastered the other steps
described here.

4. Now that each word has been pronounced in the best possible way, listen

to the total sentence rhythm and accent pattern. If it is not right, try each
of the following steps.

5. If it sounds like there should be a short pause in a particular sentence
location, but DECtalk says the sentence without a pause, try inserting a

comma between the words in question (Section 7.6.11.).
6. If the comma makes a pause that is too noticeable, try a phonemic verb-

phrase-boundary symbol instead (Section 7.6.10.).
The old man [)] stares at everyone.

7. If the wrong word is emphasized in the sentence, try to point out the word

that should receive most emphasis by placing a phonemic emphasis

symbol before it (Section 7.6.3.).

The [']Jold man is the trouble maker, not the younger one.
8. Use the pitch control symbols [/], [\], and [/\] to make final adjustments.
9. If none of these steps gives you a satisfactory sentence, you can still
specify durations and fundamental frequency motions for all phonemes
with the commands described in Section 7.7. To avoid too much trial-and-

error effort, you should have access to a speech analysis facility to analyze a recording of the way the sentence should sound.

For help with these steps, call your Digital sales representative for the names
of people who perform text optimization and exceptions dictionary
development.

143

CHAPTER 10

10.4

COMMON ERRORS

When using DECtalk, try to avoid making these two common, major errors.
1. Forgetting to change back to default voice

If you forget to return DECtalk to the standard Paul voice after using one of the
other voices, all future text will use the voice currently selected. It is a good
programming habit to return to Paul’s voice after every text message.

2. Accidental entry into phonemic mode
If MODE_SQUARE is on, permitting phonemic input, then it is possible to get
into phonemic mode unintentionally. If the text contains an unexpected [, or if

you forget to type ] after a phonemic entry, DECtalk is left in a state where it
tries to interpret text phonemically. This error makes DECtalk garble speech. In

fact, DECtalk is simply doing the best it can to interpret text phonemically,
discarding phonemically illegal letters. DECtalk keeps track of the number of

extra [ symbols in the input relative to the number of ] symbols it has seen. If
text happens to have one or more extra [ symbols, you may need to type
several ] symbols to get DECtalk out of phoneme-input mode.

144

s
=

i
G
fae
G ,"'éz’%fip“&‘@; G
o
‘”izwtvwf"n}"‘ .
L
e
.
Ce
Lo
.
L
e
w“ég"
L
. wvflfi"‘%’%’;fiv\i;
.
;
o
G
" ‘,5
.
0

L:
i

e
.

(

)

i. %@:n

G
i

;

%%3 .

‘

o
o

fi o

'q,:‘f‘

i
i o
oy
L 0

.
.
o
L
oo
‘%.
.

Ji)

Mi« o

"‘\559

o .
o
.
: .
o.
L
o
éfié‘;
:’4,\

.
.

:av
0

, g:

:.
i
g

o
s
i

z
=

o
e

i
- =

ng‘.
o

—
o

als

o
=

,»v,?,v e

|
L

oi
.
e

i
-

é&% =
-

o
-

o
o

,fi %';:

'3\'
i

‘

0
o

foa
i

At}“"%x;
SRR
. w?fi@"

S
o

;

.
.

Q:‘g

"*:\‘f‘“‘

%fl’%&*“

i
g
R

ISR

.IE: T

‘k@%“‘l;,,

..o
:

)

L
o
.
i

.
A

Qy«w

% &’s
o

%i:flf’"

=
e

i
*«fig{?
S

=
.- e

e
i
.

—

.;l

iG
L
. t‘fl}'

‘;x

sPR

- =

o 3%*“

R
-

. fi&&%‘; -

L
S

“’f}%%l..o

b
o
ol e

)

-§

f“

SREgsa

.
L
.

oo
RF{%,

i
L

= :ag

o
- =

- S
-

W"fi: i

dle
e

.
h%% @v""

. @,{d‘%@\.‘ - M'
il

wedime

)

:‘),

o =
:

éé:',==

:
i

.
e

i
:
o
.

e
e

g'w‘

e
o e
Lo
o
o
?
‘
L
@%@@%%\LV.
7% L i
w}@%
o
o
s
e
io @; : .
i L
:
;
:
.
?’ .
o
:%@A:q
; ¢ .
i
S

s
i

e @W‘f e

’%’%&%‘?«% o .
;

‘

- A

r‘°»

G
”;w,

‘

.
L

G
S

.i
.

.
.

‘

L
"44';% -.

‘

.
.

wel

.
o

fhe

.
.
.

.
..
o
o
.
o
G
:
e L
e
Go
e

.
o

- Sro

o
o

A
S

o
i

et
i

oil

e
Al
b
i vg_‘{‘g@%ri

‘

il
i

)

‘

:
H

e y?*‘)kff%ffiég%“ o

) i,

G
s

e
.

.
fiu«@w@}%«: .
o

L
.

ARy

o
.

: B

;

e
%
;

!
e

e
o

S
Sl

.
e

»"’*:?fw.

Hly

G
g

e
S

L
o

;

.
e

;

=
=

i
i

L4 .

i
C

o
ple

o
L

o
Gl

-o

.
.;

‘

@w%g
S

!
.

. .

E e

‘

.
.

. g?‘%
x

e
i

g
L

L
o
.

.
.

o
e

‘

{.(

e
o

’%’ o

L
i

oL
.

(R
-~

kfi!i:«
.

v
e

P
i
Gi

1%? 2

= :‘g;
oo

io;-

o
-

%fi}”
.

o
)

o
-

nrs

}}e-

;

.s.

-52

L
L
o

‘

DECtalk ESCAPE SEQUENCES,

STATUS REPLIES,

AND PARAMETERS

This appendix summarizes the DECtalk escape and control sequences, status
replies, and their parameters described in this manual. The following tables list

escape sequence mnemonics and their ASCII representations and parameters.
You can verify each ASCIl character by checking the decimal value that
appears below the character.

146

APPEN

DIX A

i
'

.
“:-

=
=
o

S
.

e
e

—B

:«

‘&

.
.o

= i

L
m
o

».;;-, e

.:i‘io
=

{% W

& mz:'

iki

Wil

o
et

2 anag

i
i

i
G

;

o
-

e o

.
LA
?@’3%{ o

—

o0.

.
g,

o
.

.
i
‘”,

o
e

]
"

.
2

vf;\

&
b

.
w

r
.

i'(‘a.

Hhb

e
=.

.
e

'%@’1

.
i

S
o

Seeoes o

fiy“4

.
.

ee M e
R

‘

*(;,fw

“‘l(

,g_ -

148

'7”4“:"4

. %‘%

.
e

4
o
&.‘é:fi‘ ‘Q’q? ,;j:q,\wgv

GHNR

S
S

DECtalk ESCAPE SEQUENCES, STATUS REPLIES, AND PARAMETERS

v,,

A
i

‘

o
-

e
'ifxgzw%

?ffl’:;?"

\%Ag o .

L
o

i
e

=
S-

i
.

3,.; L

.
i

Qg:&\z =

,:‘E

S
e

Chiils

=
=

e
it

.
v

- .

o
i

e
s

149

APPENDIX A
o

i
%,‘o

T
5

i ey

."% o .

D
L

ol @

o
e e
o

e
e

e
.,

i
oi
i.

i
i

.
N

0
i iy

i
Ay,

e
PN

oJons

- .

. = - - eST
.

.
G

5
eI

150

%-.-@%-WQW“.iAA.-A5\%fiw,o-=.2%,e-o--S-e-SA-«&‘-S%=oR-.3fi%@s%&%m-2=.
io

152

"
4
Gl
o

;

o 5

-- = = e - -~ e

%@m

- .,: =,, .:.

-.e.%,e=5,T=O%.-=R-.s%-o4S.es5. -;.s--X:-.i=-e=--=,o--=.=.=5.-=-E‘i-,=-o .e,s-l=S :-.::---2--oe-=V--2 IR-N-.=-=.% i£-@e-:-.:- 1)W-¥-= ...--‘.-.,e-‘-.‘s.--A.-ls.,.‘f >...e.--e‘. ...-, ‘ .fi@»e-.-- ‘..--,-‘5.=-,.%mw;&.«%.‘.-W@Wfl.%&%.,A%fi..e mW&mw%mfwiwf%?,

=--.--%Mmw.=.»wfi=w-‘--M&ex.;»f.i-%y-m%%wf%M-if:Wiw»W«%enmW@e..w-m.fMlfWemWswM-.Ws-K..W%.M;&%.W..-.-.--.-.--AoY-=fl-‘.==_A-:‘s.w.%5:&.Lm;uw-n..-s.‘,.%‘ee.-.=.o-s=--;:-.-‘‘u-%,m.w5... =‘,-m%Sw-%.\flyu.m?=s ,%.==--..-,:-fl.-.=-%os,..-|->==‘-:‘o-~-&-==.WL%=-«M,%.uemo-: wem=a5.--...>-.‘.e=:e...;:e>..-.‘--i=;o.]:.,1<:e...o
|-

. o

-=.‘.: So=e—s->=-=>>

l&;

APPENDIX A
e

i 4«‘@=

,«’

&- -

-e.E‘:

sty

ooy,

sty

CONFIGURING
THE DECtalk MODULE

This appendix shows how to configure and test a DECtalk module (C5005).

B.1

SETTING THE CONFIGURATION SWITCHES

The DECtalk module has an eight-position DIP switchpack. See Figure B-1 for
the DIP switch settings that determine the required serial line speed, data

format, and keypad mask power-up status.

If necessary, change the switch settings according to the switch position lists in

Sections B.1.1 through B.1.4.
CAUTION:

Use a ballpoint pen, a small screwdriver, or the equivalent to set

the switches. Never use a lead pencil.

The default (factory) switchpack setting provides for 9600 baud, 7 data bits,
ignore parity, and the keypad mask set to all 0s. In the default setting, all

switches are off as shown on Figure B-1.

NOTE:

154

Switch positions are on (closed) and off (open).

CONFIGURING THE DECtalk MODULE

SWITCHES
12346678

ol T TTIT
DEFAULT SETTING

|SRRERN—

(i)

——

n;u

13
pos——

MA-0347.-85

Figure B-1

DIP Switchpack Settings

1565

APPENDIX B

B.1.1

Line Speed

Switches SW6, SW7, and SW8 set the data speed on the EIA RS232-C line as
follows.

SW6

SW7

Off

9600

Off

4800

Off

2400

Off

1200

Off

600

Off

300

Off

150

110

NOTE:

Sws8

Function (baud)

DECtalk uses one stop bit at all speeds except 110 baud, when it uses

two stop bits.

B.1.2

Data Format

Switches SW4 and SW5 are used to set the data format on the EIA RS232-C
line as follows.

SW4

SW5

Function

Off

7 bits, ignore parity

7 bits, odd parity

Off

7 bits, even parity

Off

8 bits, no parity

B.1.3

Keypad Mask

Switch SW3 determines the power-up default setting of the keypad mask.
M

SW3
Off

All Os at power-up

All 1s at power-up

NOTE:

156

Function

Switch SW2 is not used.

CONFIGURING THE DECtalk MODULE

B.1.4

Busyout

Switch SW1 determines if the DM pushbutton on the front panel is enabled.
SW1

B.2

- Function

Off

DM disabled

DM enabled

TESTING THE MODULE

To test a module, perform the routine self-test as follows.

1. On the module you are checking, press the DL, CL, and ST pushbuttons
in that order (Figure B-2).

The LEDs on the module light for about 0.5 seconds, then go off for 1
second, and the SP LED flashes a few times.
2. If the module passes the test, the OK LED lights continuously.
If the module fails the test, the LEDs flash or stay off. Call Digital's Field

Service.

3. The module repeats this test every 15 seconds as long as the ST push-

button is down. To stop the test, release the ST, DL, and CL pushbuttons
in that order. |If the module passes the test, it comes up ready to operate.

NOTE:

Remember to release the ST, DL, and CL pushbuttons or the module

will continue to run the self-test.

157

(

PUSHBUTTONS 1

= OKAY
B

LEDS

[0 00 0]

APPENDIX B

= SPEAKING
= RING INDICATE
= OFF HOOK

= BUSYOUT

(@ DL

= DATA LOOP

©@cL

= CONTROL LOOP

©sT

= SELF-TEST

= EXTRA 1

1@ x2

= EXTRA 2

Tol
o

To
o)

ol
o

o
el

o)
19

o}
i

ol
©

RETAINER COVER
MA.0359.85

Figure B-2

158

DECtalk Module Controls

ABBREVIATIONS AND ACRONYMS

This appendix describes how DECtalk processes numbers, abbreviations, and
acronyms, and how it decides whether a word is pronounceable. It also
includes suggestions for correcting spoken output problems.
C.1

TEXT PROCESSING RULES
|
DECtalk processes text to be spoken by applying the following rules in this

order.

1. The input text stream is broken into groups of letters delimited by whitespace characters (spaces, tabs, or carriage returns).

2. If the letter string is not already phonemic text and is to be converted, any
understandable numbers are first expanded to their word equivalents.

159

APPENDIX C

3. Some abbreviations are expanded to their full-word equivalents. DECtalk
uses a list of numeric abbreviations and rules for a few special cases. The

user dictionary cannot override this conversion.
4.

Each letter string is broken into pronounceable entities. Punctuatio
n
(including parentheses and quotation marks), hyphenated words, and
sequences that must be spelled out are analyzed. Some abbreviations

and acronyms are recognized, plus any entries from the user dictionary.

5. Any text that DECtalk recognizes as unpronounceable (for example,
a
sequence of letters containing no vowels) is spelled out.
A few rules operate on sequences of words. Interspersing phonemic symbols
or DECtalk commands (certain escape sequences, for example) will block
these rules. Therefore, make sure that spoken text is as contiguous as possi-

ble and keep breaks in structure (from English spelling to phonemic transcrip-

tion) to a minimum.

The following terms are used in this appendix.

Character

Any of the printable ASCII characters, including letters,
digits, and punctuation.

Digit string

A string of digit characters (0 through 9). DECtalk

decides whether these should be pronounced as numbers or independent characters.
Number

A string of characters (containing digits) that are
processed as a group by DECtalk.

For example, “123” is pronounced ‘‘one hundred and
twenty-three,”” while ““1(2)3"" is pronounced ‘‘one left-

parenthesis two right-parenthesis three.”

160

ABBREVIATIONS AND ACRONYMS

C.2 NUMBER PROCESSING

DECtalk recognizes seven general number classes, and a large number of
special cases and subclasses. The general classes are as follows.

Part numbers

Strings of mixed letters, digits, and the - and /
characters.

Cardinal numbers
~

The simple numbers that are used in counting. Examples include ‘123, 123,456, ““12.345," *“01234,”
“+1.2E-4,” and ‘“12%.”

Simple strings of numbers with “st,”” “nd,” “rd,” or

Ordinal numbers

‘“th’” added, for example, “1st’”’ and “23rd.”

Fractions

Examples are “1/2,” “'2/3,” and “44/100%."

Money

Recognized by the presence of a dollar sign ($) or a
pound sign (£) as the first character of a word.

Dates

Time of day

In Digital’s standard format (23-Sep-1983), and

expandable into their English equivalent.

In the 24-hour format used by the VAX/VMS operating
system (11:04:03.02), and is spoken in its English
equivalent. The words “AM” and “PM" are correctly
processed after time values.

C.2.1

Part Numbers

A part number is defined as a string of mixed letters, digits, and the - and /

characters, containing at least one digit. The following are examples of part
numbers.

DTCO03-AA
VAX-11/780

54-15966-01
DECtalk first attempts to find the part number in the user and built-in dictionaries. If it is unsuccessful, DECtalk breaks the part number into strings of letters,
strings of digits, and separators.

161

APPENDIX C

A series of words separated by / is spelled out. DECtalk correctly speaks part
numbers of the format XXX/YYY. However, it will not recognize words such as
“Raleigh/Durham” unless they are stored in the user dictionary.

A string of digits within a part number is spoken as follows.
1. If the digit string begins with 0 or is more than nine digits long, it is spelled

out (“VS01"" becomes ‘‘vee ess zero one’’).

. One- or two-digit strings are spoken as normal cardinal numbers (‘*‘PDP11" becomes “pee dee pee dash eleven”).
. Three- or four-digit strings that end with 00 are spoken as normal cardinal

numbers (“VT100" becomes ‘‘vee tee one hundred”).

Other three-digit strings are spoken as ‘digit, pair of digits” (‘*VT220"
becomes ‘‘vee tee two twenty”).
. Other four-digit strings are spoken as ‘‘pair, pair’’ (‘“DEC 2040 becomes

“deck twenty forty”). Note that if the second pair begins with 0, it is
pronounced “zero” (“IBM 1401” becomes ‘‘eye bee em fourteen zero
O‘ne’,)'

An alphabetic string is spoken as follows.
1. One- or two-character strings are spelled out (“VT100”’ becomes ‘‘vee tee

one hundred”).

2. Longer strings are searched for in the user and built-in dictionaries. If they

are not found, they are spelled out (“DEC 2040 becomes ‘‘deck twenty
forty’).

DECtalk cannot handle all possible part numbers perfectly. The following
examples of part numbers are inconsistent with DECtalk’s number and text
processing algorithms.

162

ABBREVIATIONS AND ACRONYMS

CICS/VS

Not a part number - no digits.

net10000

“Net” is spelled out since it isn’t in the dictionary.

1E-14

DECtalk thinks this number is scientific notation.

When processing numbers and number words, DECtalk first removes leading

and trailing punctuation. DECtalk translates “(123)” as ‘‘one hundred and
twenty-three.”

C.2.2

Cardinal Numbers

A cardinal numberis a string of digits. If commas are mcluded they must break
numbers into groups of three. For example, “123,456"" is correct, but
“1234,56" is not. The latter will be spelled out as “one two three four comma
five six.”

Cardinal numbers may also include decimal fractions (‘*12.34"") and scientific
notation (‘“12.34E56"). In scientific notation, the exponent must be less than
100.

A cardinal number preceded by + or — will be spoken as “‘plus’” or “minus”
whether or not MODEWMENUS is on (Section 4.;6.2).

If the first dagatis 0 (''01234""), the number will be spoken as a string of dnglts as
would be appropriate when reading postal zip codes.

If the number is greater than 999,999,999, it will be spoken as a string of digits
with pauses between each group of three digits. If commas are provided, they
will control the pause behavior. If not, the output will pause after each group of

three digits, provided six or more digits remain. Therefore, 12345678901

will

be spoken as ‘123, 456, 78901 rather than “12,345,678,901.”
Four-digit numbers without commas are spoken in a variety of formats. For

example, “5000’’ becomes ‘‘five thousand,”’ while ‘1984’ becomes ‘‘nineteen
eighty-four.” This yields reasonable behavior when processing years.

163

APPENDIX C

Sometimes DECtalk does not understand the text well enough to pronounce
the number correctly. Here are some examples.
e The telephone number *(617) 493-8255"" will be spoken as ‘“‘six hundred

and seventeen, four ninety three dash eight two five five.” You can correct this by using one of the following steps.

Spell out the digits as ‘‘six one seven, four nine three, eight two five five”
(notice the commas to make DECtalk pause at appropriate places).

Separate the digits with spaces and commas: “617,493,8255.”
e The software cannot easily distinguish between ‘‘dash’” and “minus.”

How much is 10-15?

Bake this 10-15 minutes.
The MODE_MINUS option mode (Section 4.6.2) determines whether the

- is pronounced as ‘‘dash’ or ‘“‘minus.”

Some number formats are difficult to recognize out of context. For example,

the International Standard Date format (83.09.20) and the United States telephone number format (noted previously) are sometimes used by manufacturers
for part numbers. These ambiguous formats are not recognized by DECtalk
and you must correct them.

After a cardinal number, DECtalk recognizes a set of standard numeric abbreviations that are expanded to their English equivalent. These abbreviations are

hardwired into DECtalk and cannot be modified by the applications program-

mer.

DECtalk correctly generates

singular and

plural

forms of these

abbreviations.

Table C-1 lists the numeric abbreviations recognized by DECtalk after a cardi-

nal number. You can write them in either uppercase or lowercase letters, but
you must follow them by a period, as the table shows.

Other abbreviations, such as ‘“‘cc.,”” are spelled out by DECtalk. The period that
follows such an abbreviation is not pronounced (‘‘cc.” becomes ‘“‘see see’’) but
terminates the clause, while the period in number abbreviations does not terminate the clause.
164

ABBREVIATIONS AND ACRONYMS

g. }:“ i

)

.
e

o ;‘f«j

M,,“”

o igj«fi e
.

;

—

s
=
==

ey
i

b
.

.
.

, %t“

‘

9"3!4 )

fl;vegf;»‘i:

o
o

o
e

-AR
-

oe
--

=
-

o
e

4 ffi%}w

s
.

7
A
e

»fi

o
LA

;

A fl'h;'

i,
i

e
o

)
i

.
e

.
.

e
%

’\\g"?;. o

.S
}@"
i

e
i

e
.

.
o

b
.

e
-

L0s

b
e

=e
=

3‘:‘

s o

165

APPENDIX C

C.2.3

Ordinal Numbers

Ordinal numbers are formed from a string of digits (that may contain appropriate commas) followed by “st,” “nd,” “rd,” or “th.” Ordinal numbers are also

generated by DECtalk when fractions and dates (in standard Digital format) are
processed.

DECtalk requires that the word portion of the ordinal number be correct. For

example, “1st” will be processed correctly, but *‘2th’’ will not.

C.2.4

Fractions

Fractions consist of one or two digits in the numerator, the / character, and one

to three digits in the denominator. The numerator may range from 1 to 99, while
the denominator may range from 1 to 100. DECtalk correctly generates singular (1/3) and plural (2/3) forms.

Fractions can also use the ordinal abbreviations described in Section C.2.3, for

example, “2/3rds.”

C.2.5

Money

DECtalk assumes a digit string is money when it is introduced by the currency

symbols $ or £.

When the § or £ is recognized, DECtalk allows two forms of number strings.
e General digit strings have optional decimal fractions (‘‘$1 2.345").
e

Digit strings are

(“$12.34").

dollars and

cents (or pound

and

pence) format

DECtalk recognizes a number of quantity words (hundred, thousand, million)
that modify number processing if they immediately follow the money word. For

example, “$1.23" million is pronounced “‘one point two three million dollars.”

166

ABBREVIATIONS AND ACRONYMS

C.2.6

Dates

DECtalk mcogmzes dates written in Digital’s standard date format, such as

“23-Sep-1983," “23-Sep,” or ‘23-Sep-83.”" However, it will promunce “Sep.
23, 1983 as ‘‘September twenty-three, ninteen eighty-three.”
C.2.7 Time
of Day

DECtalk recognizes the time of day when written in the format used by Digital
operating systems. Because this format can easily be confused with part num-

ber formats, DECtalk does not try to convert the digit string. Instead, it speaks
the string with appropriate punctuation. Therefore, ““12:00"" becomes “‘twelve,
zero zero,’”’ rather than “twelve noon.”

DECtaIk‘cormctly processes VAX/VMS time values, including the fractional
second value when it is present.

C.3

ABBREVIATIONS
DECtalk recognizes, expands, and selects from a set of abbreviations taking
into the account that the abbreviations may be ambiguous.

C.3.1 Abbreviations Processed by DECtalk
In addition to the abbreviations that are recognized only after cardinal numbers, DECtalk recognizes two special cases, “‘Dr.”” and “St.”” The pronunciation of these abbreviations depends on whether the next word is capitalized.
e If the next word is not capitalized or if there is no next word (the clause

has ended), then “Dr.” is pronounced “drive”” and “St.” is pronounced
“street.” The next word must be on the same input line for the rule to
work correctly.

e If the next word is capitalized, then “Dr.” is pronounced “doctor” and
“St.” is pronounced ‘‘saint.”

Following these rules, DECtaIk correctly pronounces “Dmctor Dobbs Drive”
and ‘‘Saint Louis Street’’ in running text.

167

APPENDIX C
s,

C.3.2

Abbreviations in the Built-In Dictionary

Table C-2 lists the abbreviations stored in the built-in dictionary. Some of these
s

abbreviations are also in Table C-1.
If the abbreviation can be recognized by DECtalk during number processing,
then the English text form of the abbreviation is spoken. Otherwise, the built-in

dictionary form is spoken. Abbreviations stored in the user dictionary override

those in the built-in dictionary. The numeric abbreviations can be blocked by
including a dummy phonemic string, for example, ‘1 [ ]ft. 3.”
s

Dictionary entries that contain only uppercase letters, match text words that
contain uppercase letters in the same positions. However, the entries that

contain only lowercase letters, may match text words that contain either lowercase or uppercase letters in the same positions.

“Apr.” matches “APR.” but not “‘apr.” This is necessary to distinguish

between words at the end of a sentence and valid abbreviations, such as
“mar’”’ (to damage) and ‘““Mar.” (for March).
o

If a word in Table C-2 is written with a terminating period, you must include that

period in the input text. (It will not terminate the current clause.

[

95
i
o
&

‘

T
4

e
&%

sy,

ya3R ko
Judt

s e

Pl"

S
G
o

;

.
A. i b

)

.~
.

;
G

i
o
Wi

)

W
R

.
i

T
e
% S.:
1

b
e

y
i

e
i

i uo

i
1

Shs

o ~.} .
=

e L

s
.

?'.m& " y, A

-- —
.o

9
i

ABBREVIATIONS AND ACRONYMS

it-

i
o

L
o
o

i afi,‘éfi
o

i
[

-o

e «;'2.-’. o

o
i

.
-

.
o
o

.
.

e
oe

S-=z

e
e

wx{?’o

G
o

.
'«%‘\’m;ica:%
L B

-i

o
Wfiu

-s

oo
;
el W i %/

o
%%«

L=
-

-o

T :g’fi i

B
loeiE

e «fi%}%’; i

sty

e
P
ii

,;,w

% o

i
:

e
-i-

SRR

APPENDIX C
L

C.4

WORD SPELLOUT STRATEGIES

After number processing, DECtalk must decide whether to pronounce a string
of characters as a single word or a compound word, or if it must be spelled out.
DECtalk uses the built-in and user dictionaries and a series of word transformations to make this decision.

C.4.1

Word Spellout Tests

Number conversion, number abbreviations, and the ‘‘Street/Saint’’ test have all
been performed before DECtalk begins the decision tests. Punctuation has not
yet been removed.

1. DECtalk looks for the word in the dictionaries. First the user dictionary is
scanned, then the built-in dictionary. (Note that the user dictionary takes
precedence. If the word is found in that dictionary, the search stops.)

The dictionary lookup procedure involves stripping suffixes such as ““-ed”

and “-ing.” If the word is found in the dictionary, DECtalk speaks the
associated phonemic transcription.

. If the word is not found, any punctuation around the word is removed. If
present, the punctuation symbols *“ ([ « <[ are removed from the front of

the word, and the punctuation symbols ’ ) » > ] are removed from the end
of the word. The

square

brackets [ ]

are already discarded

MODE_SQUARE is on.
If some punctuation was removed, DECtalk performs a special test for
abbreviations “(Gen.)"” and embedded sentence punctuation (‘‘| went (last

year?) to school”).
Next, DECtalk looks for initialisms. (An initialism is a word written as a
string of uppercase letters that may or may not be separated by ‘“.”’) For
example, the string “APQ" is pronounced as ‘‘aye pea oh.” Other strings
with embedded periods are spelled out. Therefore, ‘‘a.p.0.” would be
pronounced as ‘‘aye period pee period oh period.”

If an initialism is recognized, the last ““.”” will terminate the clause, unless it
is followed by some other punctuation.

170

ABBREVIATIONS AND ACRONYMS

. At this point, all diacritical marks are removed.

If the word is still not found, it is examined for hyphenation (as in com-

pound nouns) and the single-quote character. A test is also performed to
make sure any word or word fragment has enough consonants and vowels. If the test fails, the word is spelled out.
This test ensures that the word does not contain embedded punctuation.

A word like ‘‘sys$system’ is spelled out.
If DECtalk decides the word is pronounceable, it processes each part of a
compound noun independently. If the word is not in the dictionary, it is
processed by the letter-to-sound rules.

If the word was pronounced, DECtalk examines the punctuation after the

word for silence or clause terminators. The punctuation marks " ) ]}
produce a brief silence (only one silence is produced, even if several

characters are processed). The punctuation marks ; : !, . ? terminate a
clause.

. If DECtalk decides that the word must be spelled out, the entire word is

spelled, including left and right punctuation. If the last letter of the word is
a clause terminator, it is considered punctuation and is not spelled.
10. A single letter, digit, or other character within quotes or parentheses is

spelled out (but the punctuation isn’'t spoken).

“A" is pronounced ['ey]

rather than ““uh.” This helps DECtalk process lists such as the following.
(a) books
(b) newspapers
11.

Brackets, parentheses, and braces act as commas, producing a clause
boundary. Therefore, parenthetical expressions (such as this one) sound
more natural.

12.

When text is spelled out, a brief pause is added after each character. This
makes it easier to transcribe text, such as part numbers.

171

ABBREVIATIONS AND ACRONYMS
i

C.4.2

Multinational Characters

DECtalk pronounces the names of all multinational characters. Note that this
does not mean that DECtalk will correctly pronounce words from other languages, such as “Tromsg,” ‘“‘dofa,” and ‘‘garcon.” For example,

‘dofna’’

would be pronounced [d’ownax], but A would be pronounced ‘‘en tilde.”

The Digital character sets are described in Chapter 4 of this manual.

172

'WORDS WITH TWO

PRONUNCIATIONS

Table D-1 provides a list of homographs. Homographs are pairs of words that
are spelled exactly the same but are pronounced differently. Usually, they are
noun-verb p@airs like insert (noun: stress on first syllable) and insert (verb:
stress on second syllable), although adjective-verb and verb-verb pairs do
occur. In earlier versions of DECtalk, the default form was always the noun.
Here, however, the default form is the more frequent form of the two. The
pronunciation of “transform’” as a verb (second syllable stressed) is more
common than that of the noun (first syllable stressed) and therefore the verb is
the default.

If DECtalk gives one pronunciation for a word in this list and you want the

alternative pronunciation, simply type a right parenthesis ) (no space) before
the word.

For example, if you type “‘bass,” DECtalk will say [b’eys] (the voice lower than
baritone). However, if you want the alternate pronunciation, you should type
“Ybass’’ and DECtalk will say [b’aes] (a type of fish).

173

APPENDIX D
Eo

.
.

* p@% f’% o “@%@ o
5

:
G

.
.

‘

s s e

—

e
.

y
ghn

L"
o

. %,Gm b

i
e

- e

y@;;’gfié’é@
o

o
psonsRca

g
=

.
W

S
v

Dot

I e

L
)

o o;“R
L

b
S

i
i

; il
c‘%; %\;@g i -

sl
&

ing

O i

*&0@3&&&‘&%
i

w
i
e

o
:

A
i
ot
e

i e el
b

Lo
s

W%@% .

olisa

fi%@{%‘:’fl 7

s m—

\‘

;

o
G

A
4

e
e

.
L

V‘gfié’wu’v o

5
o
ol

‘th

—

s
e

e
o

=
=

.
P

, \e%s; .

‘o

‘?f?
=

be e

-i

55 ’y;s‘g

e fm@

]

o
S

B
v
%

o G

]

e
e

‘

o
S

o
-

S
=

)

ol
oo

5
‘.
&

'e eS

:
.

o
T

) :

g
SR
G

i
a
A

;

RS o

AN
y

e
+

: ¢

». ir

o
|

S
3

e
e

B
e

S
L

Mo
oi

e
i

,%32.,.
-

.
e

e
.

*A\fé‘%?% .

e
e
=

R
L

S
'y

BL
s

g
e
Ag&vw
L

}

i
o

.
:

G
i

o
i

Ly
i

N
.

Jr—

3?3" S
o

WORDS WITH TWO PRONUNCIATIONS

ii
e

i
—

- =

s. .

. »&%.m :

o
i

;f@tfi@

‘

:
o

L ‘,&'*?&i% :fif’
s
i
R

AR
e

B
‘Mgg’\“
-.i

'\%fi‘w‘/ i
e
G

Tk
A
W
o

¢
“;xw
e
o

G
i.

;

i
L

o
o
|
i

s
.
%fi,

—

;

—=

.
’é/
.

.. 3@%

o
o

i e

; g"%’.‘i e.

i
e e

.
G

\%fk

.
‘

11
g % '1&9%&@9%

..
i

o
s

H‘Q‘%‘jfl\d

"
o

...
:
PR
fe

- =

DIX D

APPEN

.
%

,',.h.

- R
.

S
b
o
ol

‘

-s

o
==

Seaens

‘v'n;w,»‘ o

T
o
e

=
‘.,

S
"

.
Sa0

e -=2

= -

=
=

o
- eS
=

3%_" o

.
A

S
==

m g

L
.
.
.

o
Sa

i
B

e
-

o oS

o «;,
e

]

fi‘

R
w%}

. ,%,vf»‘

)

S ‘;fi;’;’;;%‘

.
4

“’k"

o
.

o
=

S(e

‘«g\ - S

‘w

«' i

]
s

o e

- R

é?;

-’««4.“-

-.
=
-

- -.

¢
!

. Q‘*GW"'4 o
.

"': 7

i
%‘;%2
‘%‘}‘

ARG

176

DECtalk /DICTIONARY

Several English phonemic alphabets are widely used today. Table E-1 lists the
phonemic alphabet that DECtalk uses, together with the phonemic alphabets

from three well-known dictionaries. It also gives an example of each sound.

Some dictionaries put the stress symbol after the vowel nucleus or at the start
of the syllable. DECtalk requires that the stress symbol appears before the
vowel.

Table E-1 has been alphabetized according to the Random House Dictionary

phonemes to facilitate translation of a dictionary pronunciation into DECtalk
phonemes.

177

APPENDIX E

=e=-:s,JI.AM-veeS=an-ctSJ.%Bfl-s&-z%3S%o---2=ee==sS;=ooRS-s==-S.so2%-:7

S)eSResie=eSyo-.sS=ie2%eoe:oeSeyRNeS-::oe=S=ei,=-.- So-eS-Se.:--o-S--=eo -soces-a-=Fo-£%W»m,o5=;..-£.E:sseo

Se=iSoaSesmo-oiaSenadns o.:-S=S=eW-voeRS=ySoiE,TtAs -eo< %@fieS=a&i%osm?;:ff%-&nB=f::=os- eS2io35%>

0!-

?@' i

s
00

e
i

s i

178

pr—

S R = SR 5 X I

S=o=

DECtalk/DICTIONARY PHONEMIC CORRESPONDENCE

o X' -

e
o . I

. I‘%%

e
i

L
e

.
i

‘

W&fi

)’3

M»’;};%;
i

=
-

o
e

. .

‘W"w‘%’c‘“&‘"{
a‘k{.)é’
o

o @u?v’ -

e
-=

iin

—=

‘? L

~‘

i
=

.
i

--S

o
.o=

o
%W .
’C%W%
.

...

;

SR
-

%S
-

L
e

PR=

s
e

v
P

e
s

=
=

o
LR

G
i

o
. o

L
.

e
L

=
-

%K%;f(f o

flffi:%?\&% e

G
e
-

;

179

APPENDIX E

E.1

WEBSTER’S NINTH NEW COLLEGIATE DICTIONARY
The following transformations are required in addition to those noted in Table
E-1.

1. Phonemic representations that appear between backslash symbols in the
dictionary should instead be enclosed in square brackets to conform with
DECtalk format.
Word

Webster’'s

DECtalk

\in\

[ihn]

2. Stress marks that appear at the beginning of each syllable should be
moved to appear just before the vowel.

Word

Webster’s

DECtalk

bat

\'bat\

[b’aet]

3. Any syllable boundary symbol in a dictionary entry should be deleted from
DECtalk phonemic pronunciations. This symbol is usually redundant information. Dictionaries are often not consistent about the use of the syllableboundary symbol - sometimes using a syllable boundary to reflect deriva-

tional structure, and sometimes to reflect pronunciation. (A syllable
boundary symbol is a legal DECtalk phoneme, but overuse or misuse of it

may block application of allophone selection rules when they should

apply.)
Word

Webster’s

DECtalk

purser

\'per-sor\

[p'rrsrr]

person

\'pers-an\

[p'rrsen]

4. The secondary stress mark may be used a bit too liberally in Webster’s

pronunciations. If unsure about including a secondary stress symbol that
seems to be there simply to prevent vowel reduction, try listening to the
word both with and without the secondary stress.

180

Word

Webster’s

DECtalk

postulate

\'pas-cha-,let \

[p’aaschaxleyt]

DECtalk/DICTIONARY PHONEMIC CORRESPONDENCE

5. The biggest difference with Webster’s is the use of the symbol for the
reduced vowel schwa. Webster’'s also uses schwa to represent the

nonreduced vowel in words such as “cup.” DECtalk expects that the
stressed phoneme [ah] will be used in words like “cup.”
In DECtalk, the reduced schwa vowel [ax] sometimes sounds better if

replaced by the raised and fronted allophone [ix], particularly if the adjacent consonants are produced with the tongue tip. Trial and error can
determine the optimal version of schwa in individual cases.
DECtalk

Word

Webster’s

cup

\'kep\

\'too-bsa\

[t'uwbax]

kisses

\'kis-8s\

[K'ihsixs]

tuba

[k’ahp]

6. The final unstressed syllable in words such as ‘“‘bottle,” “button,” and
“butter’’ sounds better if the schwa-plus-consonant pronunciation suggested by Webster’s is replaced by the syllabic consonant allophones.
Word

Webster’s

DECtalk

bottle

\ 'bat-2I\

button

\'bat-2n\

[b’ahten]

butter

\'bet-ar\

[b'ahtrr]

[b’aatel]

E.2 AMERICAN HERITAGE DICTIONARY
The primary differences between American Heritage entries and DECtalk syntax is the placement of the stress mark after the syllable rather than before the
vowel, as DECtalk requires.

Again, syllable boundary symbols should generally be omitted, and some secondary stresses may have to be omitted.

181

TEXT TUNING EXAMPLE

This appendix contains the text used to demonstrate DECtalk at the 1984 IEEE
International Conference on Acoustics, Speech, and Signal Processing. The
text is presented twice, the first time as originally written, and the second time

after phonemic and textual fixes were applied.

Original Version

[:np]
R,

A California Shaggy Bear Tale
for Seven DECtalk Voices
By Dennis Klatt
[:nh] This story was used to demonstrate DECtalk at ICASSP-84, in May of
1984, at San Diego, California.

[:np] Once upon a time, there were three bears. They lived in the great forest,

and tried to adjust to modern times.
[:nh] I'm papa bear. | love my family, but | love honey best.
[:nb] I'm mama bear. Being a mama bear is a drag.
[:nk] I'm baby bear and | have trouble relating to all of the demands of older

bears.

[:np] One day, the three bears left their condominium to search for honey.

While they were gone, a beautiful young lady snuck into the bedroom through
an open window.

182

ARG

TEXT TUNING EXAMPLE

[:nw] My name is Whispering Wendy. My purpose in entering this building
should be clear. | am planning to steal the family jewels.
[:np] Hot on her trail was the famous police detective, Frail Frank.

[:nf] Have you seen a lady carrying a laundry bag over her shoulder?
[:np]

A woman kneeling with her left ear firmly placed against a large rock

responded.

[:nu] No. No one passed this way. I've been listening for earthquakes all
morning, but have only spotted three bears searching for honey.
Changed Version

[:np]
Add periods after the title and author.
A California Shaggy Bear Tale
for Seven DECtalk Voices.
By Dennis Kiatt.

Make phonemic corrections.

[:nh] This story was used to demonstrate DECtalk at ['aykaesp] 84, in May of
1984, at [s’aendiy’eygow] California.

[:np] Once upon a time, there were three bears. They lived in the great forest
and tried to adjust to modern times.
Add emphatic stress.

[:nh] I'm papa bear. | love my family, but | love ["]honey best.
[:nb] 'm mama bear. Being a mama bear is a drag.

[:nk] I’'m baby bear and | have trouble relating to all of the demands of older
bears.

183

APPENDIX F

Begin a verb phrase.
[:np] One day, the three bears [)] left their condominium to search for honey.
While they were gone, a beautiful young lady snuck into the bedroom through

an open window.
[:nw] My name is Whispering Wendy. My purpose in entering this building

should be clear. | am planning to steal the family jewels.
Begin a new paragraph.
[:np] [+] Hot on her trail was the famous police detective, Frail Frank.
[:nf] Have you seen a lady carrying a laundry bag over her shoulder?
S

Add commas for phrasing.

[:np] A woman, kneeling with her left ear firmly placed against a large rock,
responded.

Add pitch control and emphatic stress.

[:nu] []No. No [/lone passed this [/\]way. I've been listening
for’[’]earthquakes all morning, but have only spotted three bears searching
for honey.

184

'BASIC PROGRAM EXAMPLE

DT3SAMPLE .BAS

TITLE:

VERSION:

DATE:

V1.0

3-0CT-1985

FUNCTIONAL

DESCRIPTION:

This is a sample program to control a DECtalk DTCO3.
This program is written in a "compatible'" subset
of VAX BASIC.
It should run under most BASIC implementations with
minor changes. For example, VAX BASIC no longer
requires the use of line numbers, and they have notl
been used here except as necessary to the code.

only

‘Some of the "extended" features such as the SELECT
and DECLARE statements have been avoided to make this code
‘as

transportable

possible.

‘This program has been
FORMAL

PARAMETERS:

PRIVILEGES

EXTERNAL

with VAX-BASIC V2.4.

None

REGUIRED:

None

ROUTINES CALLED:

MODIFICATION

tested on VMS V4.2,

None

HISTORY:

185

APPENDIX G

ERROR

"Dynamic

Storage,

Function,

DA$

ESC+"[c"

DASTATS

ESC+"[782;2;3c"TM

DSR$

ESC+"[n"

DSRFIRSTS$

ESC+*"[?21n*"

DSRAFTERS

DECSTRS$

and

Constant

ESC+'"[?20n"TM

Define

See

Chapter

the

descriptions

the

diagnostics
S

and

the

their

DECtalk

these

DTEND$

HzM+ESC+"\"

DTANSWERS$

DTPHONES$+"10"+DTENDS$

DTHANGUPS$

DTPHONES$+"11"+DTENDS$

DTKEYPAD$

DTPHONES$+"22"+DTENDS$

DTNOKEY$

DTPHONE$+"21"+DTENDS$
DTPHONES$+"412"

DTWINKS

DTPHONE$+"50"+DTENDS$

DTNOWINKS$

DTPHONES$+"S51"+DTENDS$

Define

The

sequences

the

take

another

parts

telephone

for

management

timeout

parameter,

them

and

ESC+"P;70;"TM

DTREPLY$+DTENDS$

PHOFFHOOKS$

DTREPLY$+"1"+DTENDS$

PHTIMEQUTS$

DTERPLY$+"2"+DTENDS$

PHTOOLONGS

DTREPLY$+"“3"+DTENDS$

PHWINKS

DTREPLY$+"4"+DTENDS$

PHNODIALS

DTREPLY$+"S"+DTENDS$

PHBUSY$

DTREPLY$+"6"+DTENDS$

PHNOANS$

DTREPLYS$+"7"+DTENDS$

the

replies

DTMASKS$

Define

the

=
=

the

number

only

that

phone

the

both
fixed

DECtalk

will

send

back

commands.

ESC+"P;83;3072*"+DTENDS

sequences
"#'"

turn

and

off

characters.

ESC+"P;10"+DTENDS

ESC+"P:;11"+DTENDS$

DTSPEAKS$

the

Define

the

and

=
=

sequences.

dial

ESC+'"P;83;0"+DTENDS$

"*"

DTSTOPS
DTSYNCS$

answers
to

escape

here.

PHONHOOK$

for

ARGy

define

DTREPLYS

DTNOMASKS$

Manual

DTPHONE$+"30;"

Define

Owner‘’s

ESC+"P;60;"

DTDIALS

replies.

DTC03

sequences.

DTPHONES

Declarations"

ESC+"[!p"

DTTIMEOUTS$=

186

19000

ESC+"P:12:;1"+DTENDS$
speaking

control

sequences.

keypad

masking

for

BASIC PROGRAM EXAMPLE

BEGINNING OF MAINLINE CODE
1000

INPUT "DECtalk Device (e.g. TTA1: or TT2:)>";DTDEVICES

DTDEVICES = EDIT$(DTDEVICES,-1%)

IF LEFTC(DTDEVICES,2%)<>"TT"

GO TO 1000

DTDEVICE$=DTDEVICES$+":" IF RIGHT(DTDEVICES ,LENC(DTDEVICES$))<>":*"

DTDEVICES$ = DTDEVICES$+“DECTALK.OUT"
OPEN DTDEVICES AS FILE #1X%

Set DTDEVICES to the string required to "OPEN" the
RSX requires a full
DECtalk device on a channel.
filespec, while VMS doesn’t. The above string works

!t
!
'

for both operating systems.

1010

UNTIL

Open the DECtalk channel.

0X%
1%

WAIT

PURGEFLAGXZ =
GOSuUB 10000

! Purge the input buffer of any possible remaining “winks" or
Keep doing it until there
other garbage characters.
!
is nothing in the buffer and BASIC returns a timeoutl error.
!
Let the error handler send control back to the next line.
!
Use a "“flag"” to tell the error handler where ito resume.
!
1020

PURGEFLAGX

WAIT 30X%

PRINT #1%, DECSTRS$

DTMASKS
DSRS$

PRINT #1%,
PRINT #1%,

!
!

Now do a "soft" reset on the DECtalk,
turn on

the

keypad mask

feature,

and request a Device Status Request to be sure the DT works.

GOSUB

10000

GO TO 32766

IF REPLY$

<> DSRS$

! Get the first response which is simply the DSR escape sequence.
GOSUB

10000

GO TO 32766 IF REPLY$ <> DSRFIRSTS$ AND REPLY$ <> DSRAFTERS
!

!
'
!

Now get

the extended DSR reply response.

This program doesn’t care if there has been a power-failure
since the user-defined dictionary isn’t being used,

so either

#1%,

GOSUB

10000

GO TO 32766

response

is OK.

DAS$

IF REPLY$

<> DASTATS

Send a Device Attribute Request to DECtalk.

Exit if it

doesn’t answer with a "DECTALK3" reply.

187

APPENDIX G

MAIN

1200

LOOP

#1X%X,

GOSUB

10000

32766

Hang

DTHANGUPS

the

#1%,

GOSUB

10000

32766

REPLY$
phone,

1210

REPLY$

for

the

"on-hook"

response.

PHONHOOKS

which

means

Waiting

for

call

at:

*;DATES$CO0X)
the

phone,

that

DECtalk

and

look

has

for

processed

“on-hook"
the

command.

600

GOSUB

WAIT

10000

ANSFLAGX

1220

look

auto-answer

answer

ANSFLAGZ

and

"XDT3SAMPLE-I-WAITFORCALL,

Enable

WAIT

PHONHOOKS

DTANSWERS

TIME$COX);*

32766

REPLY$

PHOFFHOOKS

Reset

the

answer

Then

reset

Use

ANSFLAGX

re-run

This

activity

Therefore,

greater.

systems.

the

VMS

#1%,

10000

32766

timeout

phone.

The

the

timeout

tell

diagnostics

will

GOSUB
TO

input

the
BASIC

This

The

handler

phone

isn’t

bad

minute

allocates

always

sets

and

wait

response

for

DECtalk

"off-hook."

back

seconds.

between

only

works

error

the

line.

RMS

mins.,

proper

the

distinguish

since

DECtalk

for
of

255

minutes.

doesn’t

the

different

and

within

and

timeout

byte

timeout

differently

answered

work

parameter.
when

sees

anything

operating

DTKEYPADS$

REPLY$

Enable

the

telephone

mode

the

DECtalk3

whenever

sees

PHOFFHOOKS

keypad

does
a

DTMF

the

new

"internal"

tone

(key

"Auto-stop'"

stop/disable

mode.

this

speaking

press).

188

BASIC PROGRAM EXAMPLE

1230

#1%,

DTSPEAKS

#1%,

"[_<100>]Hello,

#1%,

the

#1%,

multi

“"control

#1%,

"Now

would

1240

#1%,

10000

32766

the

REPLY$

Enable

and

get

the

command

detector

look

"Wink"
the

#1X%,

"Press

#1%,

telephone

speaking."

DECtalk

synthesizer
lines

demonstrate

commands

the

entered

family."

that

can

simultaneously."

ability

from a

push

user
it

button

Phone."

hangs

the

phone.

should

"offhook"

PHOFFHOOKS

detector
is

10000

GO TO

1010

the

keys

then

hang

REPLYS$

keys

does

REPLY$

"*"

the

OK.

the

key

and

end

implied
do

phone

Don’t

enable
the

the

number

PHWINKS

pressed

the

picking

followed

an
to

DECtalk

the

phone

wink

can

DECtalk.

back

star

thing

the

read
the

mode

first

see

from

processed

any

"Press

GOSUB

Get

the

voice

response

like
a wink

"Auto-stop"

line

until
now because

“will
PRINT

DECtalk

DTWINKS

member

“"understand

GOSuB

this

latest

number

you

this

have

key,

and

demonstration.TM

"disable

sign

entered."

speaking"

re-enable

it.

command,

Then

speak.

it’s

look

for

and

start

“wink."

“wink"

again.

THEN

#1%,

DTSPEAKS

#1%,

"Thank

#1%,

"Have

#1%,

DTSYNCS$

1200

#1%,

DTSPEAKS

#1X%X,

LEFTC(REPLYS$,LENCREPLYS$)-1%);"."
DTKEYPADS$

you
a

for

nice

calling

DECtalk

3."

day."

ELSE

END

!
!

#1%,

GOSUB

10000

32766

1240

REPLY$

PHOFFHOOKS

Speak

message

otherwise

read

and

hang

back

the

phone

number

and

if
go

"“*"
get

pressed,

another.

189

APPENDIX G

DEFINE

10000

REPLYS$

FUNCTION

GET

INPUT

FROM

DECTALK

"@u
R

10010

INPUT

LINE
IF

#1X,

SEQS

RIGHT(SEQS$ ,LENCSEQ$)-1X)=CR+LF

THEN

REPLYS$

REPLY$+LEFT(SEQS$,LENC(SEQG$)-2%)

REPLY$+SEQS$

ELSE
REPLYS$
GO

END
!

10010

BASIC

Since

<ESC>

implemented

character

around

"keypad

until

look

get

the

VMS

<CR>

mask' mode.

for

and

terminator,

that

BASIC

and

also

strip

RSX

keep

thinks

that

the

looping

DECtalk

appends

translates

<CR>

output

<CR><LF>,

off.

RETURN

GLOBAL

19000

19010

RESUME

1010

ANSFLAGX

RESUME

1020

PURGEFLAGZ

Got

timeout
the

waiting

timeout

came

response

from

expected

place,

the

line

and

hang

the

phone.

code.

"XDT3SAMPLE-F-BASICERR,

32767
it

END

from

back

response

for

190

DECtalk."TM

1200

any

other

error,

invalid
Print

escape

response
error

then

sequence

message

then

Otherwise,

message

and

exit.

escape
and

DECtalk.

";ERR,ERL,ERT$CERR)

“XDT3SAMPLE-F-INVESCSEQ,
“"Invalid

32767

«<>15%

!
32766

ERR

"XDT3SAMPLE-F-NORESPONSE,

RESUME
!

TRAP

RESUME

19010

ERROR

received"

sequence

exit.

was

received.

gy,

SERVICE AND DOCUMENTATION

H.1

DIGITAL’S SERVICES

Your DECtalk DTCO03 system comes with a warranty that gives you access to
the best of Digital's resources, including our technical expertise, spare parts
inventories, and worldwide service organization.

After warranty, Digital’s support continues through a range of on-site and off-

site service alternatives for as long as you own your Digital hardware. These
services are designed to help you get the maximum performance from your
system.

H.1.1

DECservice

DECservice is Digital’'s most comprehensive on-site service. It provides committed response time including a four-hour service response if your system is

within 100 miles of a Digital service location. DECservice also provides continuous repairs until the problem is solved, a preventive maintenance program,
installation of the latest engineering changes, and automatic escalation for
complex problems.

DECservice
lets you choose the hours of coverage you need to support your

application, up to 24 hours a day, 7 days a week.

191

APPENDIX H

H.1.2

Basic Service

Basic service offers economical yet full service coverage. Your calls for service

receive priority, second only to DECservice calls. You also receive preventive
maintenance, installation of the latest engineering changes, and automatic

escalation of complex problems.

Basic service is available during standard business hours, from 8:00 a.m. to
5:00 p.m., Monday through Friday.
H.1.3

Per Call Service

If your application does not demand comprehensive support, you can take

advantage of one of Digital’'s per call programs.
Per call service is available on-site and off-site on a noncontractual basis.

Service is available Monday through Friday during standard business hours,
from 8:00 a.m. to 5:00 p.m.
For on-site per call service, you pay for the time and materials needed for each
service call. Off-site per call service is available through mail-in board replacement and carry-in system repairs.

H.1.4

Carry-In Service

This is Digital’'s low-cost alternative to on-site support. You can carry your |
DECtalk into any of Digital’s 150 Service Centers throughout the United States
and pick up the repaired system within two days. Or, if you perform your own
maintenance, you can carry in your faulty module and receive an immediate
exchange.

Carry-in service is available through a one-year, fixed-cost agreement or on a
per call basis that includes a flat rate for labor, plus the cost of all parts used in

the repair. All carry-in service and parts come with a 90-day warranty.
g

192

SERVICE AND DOCUMENTATION

H.1.5

DECmailer Service

DECmailer is a factory replacement service for Digital’s customers who maintain their equipment to the module or subassembly level. It provides 5-day
turnaround, free return shipping, 90-day warranty, 24-hour emergency service,
monthly billing, and quarterly activity reports.

Whichever Digital service option you select, you will receive high-quality, reliable service delivered by one of the largest service organizations in the industry.
H.1.6 Digital Field Service Information Numbers
For more information on any of Digital’s services, call the service information
number in your area during normal business hours.
Australia

(02) 4125555

Austria

(222) 6776410

Belgium

(02) 2425095

Canada

(800) 267-5251

Denmark

(2) 889666

Finland

(0) 423511

France

(6) 0778292

Holland

(30) 640293

Ireland

(1) 308433

Italy

(02) 617961

Japan

(03) 989-7161

New Zealand

(09) 595-914

Norway

(2) 160290

Portugal

(1) 725402

Spain

(1) 7331900

Sweden

(8) 7338000

Switzerland

(01) 8169111

United Kingdom

(734) 868711

United States

(800) 554-3333

West Germany

(089) 95910

193

APPENDIX H

H.2