Contents / Chapter 6 / Appendix A
This chapter describes TTS and ASR configuration options of WireLess TelNet
5250.
A detailed description of voice functions is available in the xoiXtreme
for WireLess TelNet voice system Reference Guide
The Voice configuration file is a text file containing the configuration values for the TTS (Text To Speech) and ASR (Automatic Speech Recognition) options. The name of this file is always "_WTnVoiCE.cfg". You can modify its values using "_WTnVoiCE.iniedit" (configuration helper).
The Voice configuration file is a "Windows dot-ini-Like" text file, which includes:
The lines in the configuration file must be: Comments, Sections, Keys + Values or empty lines.
Sections group keys and values. Section names are in capital
letters surrounded by brackets, i.e.: "[TTS_DEF]" denotes the TTS_DEF section.
The order of the sections into the file is not relevant.
The keys are keywords that have sense only in their own section, so they can't be moved from one section into another. Case (lowercase or uppercase) is not relevant for keywords, but are more readable... Keys must be followed by "=" sign and the value of the parameter, i.e.: VoiceSpeed=70.
Values follow the keys and "=". Each parameter expects the value in a fixed format (indicated with the parameter definition like "<format>"). Values may be empty in some cases.
Formats are:
<Yes/No>
Needs Yes or No value.
<string / xxxx>
ASCII text characters, delimited or not by " character. If the
string is delimited by ", the spaces (20h) are preserved, otherwise
are ignored. xxx indicates the units or the meaning of the value.
<hex-string>
In this format, three characters represents one final character.
There are three ways to supply "hex-strings" (each target character takes
always 3 source characters):
| Mnemonic | Format | Description |
|---|---|---|
| "Escaped" | "/HH" | A slash ("/" or "\") and a two-digits hexadecimal value, i.e.: \1B for character 27. |
| "Numeric" | "999" | Three decimal digits, i.e.: 065 for character "A". |
| "Alphabetic" | "..A" | Two dots and one alphabetic character, i.e. : ..X for character "X". |
Hex-strings may be delimited by " to be more readable.
<num / xxxx>
Decimal numeric value (0 - 9). xxxx indicates the units or the meaning of the
value.
<word>
Hexadecimal four-digits value. i.e.: A50F.
<uppercase letter>
One uppercase character. i.e.: N.
[TTS_DEF]
This section groups the global Text-To-Speech (TTS)
settings, and enables the screen or printer to process voice actions.
[ASR_DEF]
This section groups the options for
Automatic Speech Recognition (ASR) TelNet voice applications.
Voice triggers are several sections that allow to perform voice actions depending on conditions to be found on the emulated screen. A voice trigger is composed of three sections:
[VOICETRIGGER_XX]
This section defines the conditions to look for in the
emulated screen to run trigger actions.
[VOICETRIGGER_TTS_XX]
This section groups the TTS actions to do when the voice trigger runs.
Constant text (supplied in the trigger) and variable text (extracted from
emulated screen) will be read to user.
[VOICETRIGGER_ASR_XX]
This section contains the ASR actions to perform in
the trigger.
VoiceSpeed=<num> (70)
Relative speed of speech (0 - 100).
TtsPrinterOn=<Yes/No> (No)
Yes=Use the voice TTS (Text To Speech) instead of printing data.
No=Do not use TTS.
MsgWelcome=<string> (Welcome to Telnet)
Message to say at startup.
MsgConnecting=<string> (Connecting...)
Message to say when connecting.
MsgConnected=<string> (Connected)
Message to say when connected.
MsgDisconnect=<string> (End of connection)
Message to say when disconnected.
KeyRepeat=<num> (2070)
Repeat last TTS message Hot Key. This is a keyboard Scan Code.
KeyVolUp=<num> (2071)
TTS Volume Up Hot Key. This is a keyboard Scan Code.
KeyVolDn=<num> (2072)
TTS Volume Down Hot Key. This is a keyboard Scan Code.
KeySpeedUp=<num> (2073)
TTS Speed rate Up Hot Key. This is a keyboard Scan Code.
KeySpeedDn=<num> (2074)
TTS Speed rate Down Hot Key. This is a keyboard Scan Code.
ReliabilityLvl=<num> (4000)
Minimum reliability level necessary to accept word recognition.
When a word is recognized, ASR engine returns a “reliability level” which is
the confidence level that the engine assigns to the recognition. A low
reliability level denotes a confusing recognition that will be rejected.
CustomGrammar=<string> (<custom>)
Name of the grammar referenced by the input mode parameter in VOICETRIGGER_ASR /
Grammar.
ResCancel=<string/command> (Cancel)
This is the keyword of the "commands" grammar the
user should say to cancel an action.
ResAccept=<string/command> (Ok)
This is the keyword of the "commands" grammar the
user should say to accept an action.
ResRepeat=<string/command> (Repeat)
This is the keyword of the "commands" grammar the
user should say to ask for a playback (voice announce before ASR) during an ASR
recognition.
SyncTts=<Yes/No> (Yes)
This option forces the ASR engine to wait until the TTS engine ends (speech by PDA) before starting user's voice recognition.
On=<Yes/No> (No)
Enables / disables the trigger.
Row_X=<num> (0)
Row position to look for a text. 0 (zero) means all rows.
Col_X=<num> (0)
Column position to look for a text. 0 (zero) means all columns.
Row=0 and Col=0 means anywhere in the screen.
Match_X=<string>
(empty)
Text to look for at Row_X / Col_X position. To consider ending spaces, text
should be delimited by quotes “”.
If no text to look for is supplied, the condition is not evaluated (TRUE).
CursorAtRow=<num> (0)
Unprotected input field row.
Min= 1, Max= 24.
CursorAtCol=<num> (1)
Unprotected input field column.
Min= 1, Max= 80.
On=No
Enables / disables TTS actions in the trigger.
TtsSay01=<string> (empty)
Combination of constant text (normal text in key) and variable text (delimited
by curly brackets {}).
Variable text is denoted by three numeric values indicating its location by row
and column and its length.
All the text is read as words, except for those parts delimited by special tags.
<SPL></SPL> special text tags delimit text to spell, i.e. the text
“location <SPL>AB8</SPL> to go” will be read out loud as
“location ey be eight”.
On=<Yes/No> (No)
Enables / disables ASR action for this trigger.
Grammar=<0=commands/ 1=digits/ 2=custom>
(1)
Grammar to use in the trigger.
A grammar is a set of rules and words that ASR engine uses for speech recognition.
Standard ASR engine includes two grammars: commands (Cancel / Ok) and digits (0, 1, 2, 3, 4, 5, 6, 7, 8, 9).
Customized grammars will be built further.
LenMin=<digit> (2)
Minimum voice input length expected. Inputs under this value will be rejected and a new input will be requested.
LenMax=<digit> (4)
Maximum voice input length expected. Inputs bigger than this value will be rejected and a new input will be requested.
Confirm=<Yes/No> (Yes)
Yes=Uses the Confirm voice input algorithm to perform ASR input.
No=Uses the Simple voice input algorithm for ASR.
Repeat=<Yes/No> (Yes)
Yes=Activates Repeat algorithm to perform ASR input.
CancelDo=<Yes/No> (Yes)
Allows user to cancel a data input and sends cancel sequence to host.
Cancel=<string> (0123)
Keyboard scan code sent to host when user cancels input without any data.
Default value, 0123 = CURSOR-UP (see TN52_CURSOR_KEYS).
CancelConfirmation=<string> (Do you wish to cancel?)
Phrase that will be said by the TTS engine before the cancel command is confirmed.
Terminator=<num / Action> (6)
Terminator action that will be performed after the ASR data to validate input.
Default value =FIELD-EXIT
'A'=F1
'B'=F2
'C'=F3
'D'=F4
'E'=F5
'F'=F6
'G'=F7
'H'=F8
'I'=F9
'J'=F10
'K'=F11
'L'=F12
'M'=F13
'N'=F14
'O'=F15
'P'=F16
'Q'=F17
'R'=F18
'S'=F19
'T'=F20
'U'=F21
'V'=F22
'W'=F23
'X'=F24
'0'=ENTER
'1'=CLEAR
'2'=PGDN
'3'=PA1
'4'=PA2
'5'=PA3
'6'=FIELD-EXIT
Empty=No action
This action has the same function that [TN52_BARCODES]
PostAction
Contents / Chapter 6 / Appendix A